Top Banner
Indicative of the vast amount of unfrozen water on northern Christmas Bird Counts in the 102nd CBC season was this Long-tailed Duck (Clangula hyemalis), a winter adult male, at Michigan City Harbor on the Indiana Dunes National Lakeshore, Indiana, CBC. Photo/John K. Cassady ON THE COVER: Hummingbirds have become a real phenomenon on Christmas Bird Counts, both in their “expected” area of abundance along the Gulf Coast and in southern areas, and increasingly on far-flung CBCs to the north. This Rufous Hummingbird (Selasphorus rufus), an immature male, was attracted not by feeders but by late-blooming flowers, and spent the fall at Lenoir Preserve in Yonkers, New York, to be counted on the Bronx-Westchester, New York, CBC. Photo/Phil Jeffrey The 102nd Christmas Bird Count American Birds The 102nd Christmas Bird Count Published by the National Audubon Society June 2002 Frank B. Gill Senior Vice-President for Science Geoffrey S. LeBaron Christmas Bird Count Director and Editor-in-Chief Kathy Dale Director of IT, Audubon Science Richard J. Cannings Christmas Bird Count Coordinator, Bird Studies Canada Ruth Helmich Avery English Christmas Bird Count Assistants Andrew Farnsworth CBC Research Associate Carol Capobianco Managing Editor [email protected] Art Director/Graphic Designer Greg Merhar Graphics Associate Caroline Jackson GIS Technician Heidi DeVos Production Director CONTENTS The 102nd Christmas Bird Count . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2 Geoffrey S. LeBaron Christmas Bird Count Canada: 2001-2002 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8 Richard J. Cannings Using Christmas Bird Count Data in Analysis of Population Change . . . . . . . . . . . .10 John R. Sauer and William A. Link Long-term Trends of Raptors on CBCs in the Midwest . . . . . . . . . . . . . . . . . . . . . .15 Kelly J. McKay, Jon W. Stravers, Peter C. Petersen, Casey J. Kohrt, Joseph S. Lundh, and Gary V. Swenson Yellowstone National Park: A Look at an Individual Count Circle . . . . . . . . . . . . . . .22 Terry McEneaney The Use of Christmas Bird Count Data to Monitor Populations of Exotic Birds . . . . .24 Bill Pranty WatchList Species as Viewed Through the Christmas Bird Count Database . . . . . . .29 Andrew Farnsworth Alphabetical Index to Regional Summaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . .32 Regional Summaries of the 102nd Christmas Bird Count . . . . . . . . . . . . . . . . . . .33 Count Circles, Codes, and Compilers in the 102nd Count . . . . . . . . . . . . . . . . . . .94 Summary of Highest Counts of Individuals for Canada . . . . . . . . . . . . . . . . . . . .104 Richard J. Cannings Summary of Highest Counts of Individuals for the United States . . . . . . . . . . . . .107 Brent Ortego Audubon Science Center 545 Almshouse Road Ivyland, PA 18974 www.audubon.org/bird Bird Studies Canada/ Études d’Oiseaux Canada P.O. Box 160 Port Rowan, Ontario N0E 1M0 www.bsc-eoc.org
32

American Birds - National Audubon Society | Audubon

Sep 12, 2021

Download

Documents

dariahiddleston
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: American Birds - National Audubon Society | Audubon

Indicative of the vast amount of unfrozen water on northern Christmas Bird Counts inthe 102nd CBC season was this Long-tailed Duck (Clangula hyemalis), a winter adultmale, at Michigan City Harbor on the Indiana Dunes National Lakeshore, Indiana, CBC.Photo/John K. Cassady

ON THE COVER: Hummingbirds have become a real phenomenon on Christmas Bird Counts, both in their “expected” area ofabundance along the Gulf Coast and in southern areas, and increasingly on far-flung CBCs to the north. This Rufous Hummingbird(Selasphorus rufus), an immature male, was attracted not by feeders but by late-blooming flowers, and spent the fall at LenoirPreserve in Yonkers, New York, to be counted on the Bronx-Westchester, New York, CBC. Photo/Phil Jeffrey

The 102nd Christmas Bird CountAmerican BirdsThe 102nd Christmas Bird Count

Published by the National Audubon SocietyJune 2002

Frank B. GillSenior Vice-President for Science

Geoffrey S. LeBaronChristmas Bird Count Director andEditor-in-Chief

Kathy DaleDirector of IT, Audubon Science

Richard J. CanningsChristmas Bird Count Coordinator,Bird Studies Canada

Ruth HelmichAvery EnglishChristmas Bird Count Assistants

Andrew FarnsworthCBC Research Associate

Carol CapobiancoManaging Editor

[email protected] Director/Graphic Designer

Greg MerharGraphics Associate

Caroline JacksonGIS Technician

Heidi DeVosProduction Director

CONTENTS

The 102nd Christmas Bird Count . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2Geoffrey S. LeBaron

Christmas Bird Count Canada: 2001-2002 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8Richard J. Cannings

Using Christmas Bird Count Data in Analysis of Population Change . . . . . . . . . . . .10John R. Sauer and William A. Link

Long-term Trends of Raptors on CBCs in the Midwest . . . . . . . . . . . . . . . . . . . . . .15Kelly J. McKay, Jon W. Stravers, Peter C. Petersen,Casey J. Kohrt, Joseph S. Lundh, and Gary V. Swenson

Yellowstone National Park: A Look at an Individual Count Circle . . . . . . . . . . . . . . .22Terry McEneaney

The Use of Christmas Bird Count Data to Monitor Populations of Exotic Birds . . . . .24Bill Pranty

WatchList Species as Viewed Through the Christmas Bird Count Database . . . . . . .29Andrew Farnsworth

Alphabetical Index to Regional Summaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . .32

Regional Summaries of the 102nd Christmas Bird Count . . . . . . . . . . . . . . . . . . .33

Count Circles, Codes, and Compilers in the 102nd Count . . . . . . . . . . . . . . . . . . .94

Summary of Highest Counts of Individuals for Canada . . . . . . . . . . . . . . . . . . . .104Richard J. Cannings

Summary of Highest Counts of Individuals for the United States . . . . . . . . . . . . .107Brent Ortego

Audubon Science Center545 Almshouse RoadIvyland, PA 18974www.audubon.org/bird

Bird Studies Canada/Études d’Oiseaux CanadaP.O. Box 160Port Rowan, Ontario N0E 1M0www.bsc-eoc.org

Page 2: American Birds - National Audubon Society | Audubon

What is it that entices otherwise sanehuman beings—even if they are birdwatchers—to spend a significant chunkof their holiday vacation time away fromfamily and friends out in the cold weath-er counting birds? Is it the lure of thepossibility of finding magnificent winterraptors, twittering flocks of irruptivefinches, out-of-range vagrant birds, orout-of-season lingering fall migrants?Could it be the hope of breaking a countrecord from last season (or even better,the record of a neighboring count)? Howabout the simple prospect of spending alovely early winter day outdoors?

Or is it actually that the ChristmasBird Count is the only time we may seesome of our friends and family?

Any one of these things—and quiteprobably a litany of other similar possibil-ities—draws tens of thousands ofobservers into the field, or to their win-dows to watch their feeders, each holiday

season to participate on the annualChristmas Bird Count (CBC). But the102nd CBC proved to have an incrediblearray of these potential observer catalysts.A mild and dry fall season set the stage forthe likelihood—and survival—of linger-ing species and out-of-range wanderers.The enticement of good numbers ofwestern hummingbirds, Ash-throatedFlycatchers, and Varied Thrushes in theeast, all dutifully reported on rare birdalerts during the fall, whetted CBCobservers’ appetites.

Additionally, it seems that the weatherconditions of the summer and fall in thenorthern regions of North America wereunfavorable for seed crops and smallmammal crops, so both winter finches (inthe form of redpolls and crossbills at least)and irruptive raptors (especially Snowyand Short-eared owls early on) began asouthward movement. And generallymild and snow-free weather conditions

continentwide kept ponds, lakes, andrivers unfrozen far north of the usual“freeze line,” enticing an incredible arrayof water-associated bird species (loonsand grebes, waterfowl, herons, and marshand shorebirds) to linger in significantnumbers far north of their usual CBCranges. To top it all off, weather condi-tions, especially on weekends, during theCBC official count period—fromDecember 14 through January 5—tend-ed to be “better than average.” By allthese measures, it seemed that the 102ndChristmas Bird Count was shaping up tobe a grand slam of a CBC.

In the final result, and by many meas-ures (especially of the humancomponent), the 102nd Count resultssurpassed every preceding season. Therecord total of 1880 counts set during lastseason’s 101st CBC was shattered thisyear. With 307 counts in Canada, 1577counts in the United States, and 52 in theCaribbean, Latin America, and thePacific Islands, 1936 CBC circles werecovered this season, up an amazing 56from the 101st Count. Not surprisingly, asignificant roster of 54 new ChristmasBird Counts is included in this total—19in Canada (including two new counts inNunavut), 31 in the United States, one inthe Pacific Islands, two in Latin America,and an intriguing new pelagic count donefrom a research vessel sailing betweenAntarctica and southern South America.That’s quite a different type of countthan Mad Island Marsh, Texas, orMonteverde, Costa Rica. The amazinggeographic scope of the Christmas BirdCount is documented even in the list ofnew counts—from the Arctic in Nunavutto the oceans surrounding Antarctica.This varied roster of new CBCs is pre-sented for your perusal in Table 1.

The observer pool attending thosecounts was equally impressive. The previ-ous record number of participants, fromthe landmark 100th Christmas Bird

2 AMERICAN BIRDS

The 102nd Christmas Bird CountDecember 14, 2001, to January 5, 2002

Geoffrey S. LeBaron

One of the defining features of the 102nd Christmas Bird Count across many areas inthe northern and central latitudes of North America was a good flight of Snowy Owls(Nyctea scandiaca). This immature bird was photographed at Jones Beach State Park,New York, in January 2002. Photo/Phil Jeffrey

Page 3: American Birds - National Audubon Society | Audubon

Count, was just over 52,000 observers(on 1823 counts). During last season’s101st Count, while the number of countsincreased, the observer pool was slightlylower—due in large part to generallyinclement weather during the CBC peri-od. That trend was again reversed (boththe weather and the participation) duringthe 102nd Count. In Canada, 7190 fieldobservers partnered with 4046 feederwatchers; in the United States, 38,287were in the field along with 5565 yard-bird observers; and in the Caribbean,Latin America, and the Pacific Islandsanother 1036 field and five feederobservers participated, for a whoppingtotal of 56,129 observers in the 102ndChristmas Bird Count.

On first impression, it seems surprisingthat no new “countable” species wereadded to the cumulative roster of birdsfound during the Christmas Bird Countin North America, and that the totalnumber of birds included is somewhatlow. It may be counterintuitive that a newrecord number of observers in a newrecord number of census areas did notcombine to tally a record number ofbirds—or any new species in NorthAmerica. But consider that one major fac-tor generating this record observereffort—mild weather—is beneficial tobirds as well as birders. Remember thatwhen environmental conditions are mild,birds are neither concentrated in areas oflimited food resources nor hastenedsouthward to warmer climes. Comfortlevel aside, the harsh winter conditionsthat can be “good” for observers (thoseproducing impressive tallies of early-win-ter birds) are actually stressful for many ofthe birds themselves. While there mayhave been many more birds than averagein North America during the 102nd CBCseason, they were dispersed across the con-tinent, and not concentrated in the primeareas covered by CBC circles. And speciesforming huge flocks during harsh condi-tions (like blackbirds and starlings) wereapparently still dispersed as well, so the51,908,813 total birds counted(47,241,040 in the United States,3,720,457 in Canada, and 947,316 in the

Pacific Islands, Latin America, and theCaribbean), while impressive, is still an“average” tally. As far as the blackbirds andtheir effect on total numbers of birds,remember that one count circle (such asPine Prairie, Louisiana, in the 88th CBC)can more than double the entire numberof birds counted during a season’sChristmas Bird Counts!

The total taxonomic tally included inthe 102nd Christmas Bird Counts inNorth America and Hawaii was 657“countable” species, 38 additional forms,and 23 “non-countable” exotic species.There were some exotic species tallied forthe first time, including Scarlet-frontedParakeet, and White-fronted, Yellow-naped, and Yellow-crowned parrots.There is an incredible diversity of free-fly-ing introduced species (especially parrots)from around the world in areas like southFlorida and southern California (not tomention Hawaii); and, as discussed inBill Pranty’s feature article “The Use ofChristmas Bird Count Data to MonitorPopulations of Exotic Birds” in this issue,the CBC is a valuable tool to monitor thestatus of these introductions. Of course,although the majority of Christmas BirdCounts each season are in NorthAmerica, the majority of species talliedare included in the 52 counts outsideCanada and the United States. This sea-son, a total of 1828 species was tallied onthose 52 counts—an impressive number,but not surprising considering the geo-graphic scope of coverage. Note thatthere is considerable overlap between the657 species on the North American listand the 1828 outside North America;many are counted in multiple regions.

A look at species totals on counts dur-ing the 102nd season provides aneducation on the effects of mild weatherin different portions of the continent.Table 2, as usual, is the list of all counts inthe CBC’s 102nd year with 150 or morespecies. In a slight variation of traditionalordering, this year’s Table 3 is the list ofhigh counts per region—species totals asreported by compilers and regional edi-tors. Continuing its reign at the top of thespecies total category in North America is

THE 102ND CHRISTMAS BIRD COUNT AMERICAN BIRDS 3

Table 1. New Counts in the 102nd(2001-2002) Christmas Bird Count

Count Name Count Code

CANADACold Lake, Alberta ABCKDinosaur Provincial Park, Alberta ABDINanton, Alberta ABNARocky Mountain House, Alberta ABRMHat Creek, British Columbia BCHCKingfisher, British Columbia BCKFKaslo, British Columbia BCKOMcBride, British Columbia BCMCPemberton-Mt. Currie, British Columbia BCPBTofino, British Columbia BCTOClarenville, Newfoundland NFCLLunenburg, Nova Scotia NSLUPictou Harbour, Nova Scotia NSPIArviat, Nunavut NUARRankin Inlet, Nunavut NURIGameland, Ontario ONGAEast Point, Prince Edward Island PEIEPQu’Appelle Valley Dam, Saskatchewan SKQDTorch Valley, Saskatchewan SKTV

UNITED STATESCullman, Alabama ALCUCrooked Creek Valley, Arkansas ARCCPhoenix-Tres Rios, Arizona AZPTJohn Martin Reservoir, Colorado COJMMilford, Delaware DEMFFloyd County, Georgia GAFCSiam-Hopkins, Iowa IASIBig Oaks N.W.R., Indiana INBOJetmore, Kansas KSJMNorth Penobscot Bay, Maine MEBFNorthern Wright County, Minnesota MNNWPine County, Minnesota MNPCTwo Harbors, Minnesota MNTHLiberal, Missouri MOLIWest Yellowstone, Montana MTWYAlligator River N.W.R., North Carolina NCARLake Ilo N.W.R., North Dakota NDLIAngel Fire-Eagle’s Nest, New Mexico NMAEPecos, New Mexico NMPCRaton, New Mexico NMRANew Woodstock, New York NYNWRedmond, Oregon ORRDCalamity Creek, Texas TXCEOrange County, Texas TXOCQuanah, Texas TXQNMorgan, Utah UTMGLeavenworth, Washington WALWHartford, Wisconsin WIHAHustisford, Wisconsin WIHUHayward, Wisconsin WIHWElkins Area, West Virginia WVEL

PACIFIC ISLANDS, LATIN AMERICADrake Passage, Antarctica CHDPCerro Blanco-Chognon-Puerto Hondo

Estuary, Ecuador ECCBEnsenada, Baja California, Mexico MXESPalmyra Atoll, Pacific Islands PIPA

Page 4: American Birds - National Audubon Society | Audubon

Mad Island Marsh, Texas, this year with aslightly lower species total of 233 than itstally in the 101st Count of 235. An inter-esting pattern develops when this year’stotals are compared with those of other,more wintry seasons. First, at the verynorthern fringe of Christmas Bird Countcoverage, there is little or no change inspecies totals from year to year. All fourArctic counts—Nome and Prudhoe Bayin Alaska, and Arviat and Rankin Inlet inNunavut—only had one species(Common Raven) on count day (thoughArviat had two others, Willow Ptarmiganand House Sparrow, during count week).This is the norm—in the entire run of thePrudhoe Bay CBC since its inception inthe 88th Count, the only species ever tal-lied is Common Raven, although the two“cw” species this season from Arviat maygive observers at Prudhoe Bay hope fordoubling—or tripling—their list in somefuture season. Moving southward mental-ly as we peruse species totals, in the 102ndCount it is apparent both from Tables 2and 3 (and from the regional summariesin the pages that follow) that many recordspecies totals were set this season inAlaska, southern Canada, and acrossmuch of the northern two-thirds of theLower 48. Moving farther south, many ofthe species totals across the southernUnited States are lower than in someyears. As mentioned earlier, factors such asthe combination of mild weather allowingsurvival of vagrants and late migrants tothe north, combined with the northwardlingering of waterfowl and shorebirds inunfrozen marshes, lakes, and rivers tendedto both increase species totals northwardand decrease them in the south.Comparing this season’s Table 2 (150+species) with that of other years, there isan indication that species totals in themid-Atlantic region and even in BritishColumbia were high, and thus appear in thistable, while the totals of some counts farthersouth in California, Texas, Louisiana, andFlorida that may tally 185 species or moreare in the 150s to 170s. So it is possiblethat major changes in global weather pat-terns, such as pre-“El Niño” seasons andany potential human-induced climate

4 AMERICAN BIRDS

Table 2.Counts with 150 or more species recorded on the 102nd (2001-2002) Christmas Bird CountTable 2a: Counts north of the United States–Mexican border

Count Code Rank Count Name Species RecordedTXMM 1 Mad Island Marsh, TX 233CASB 2 Santa Barbara, CA 206TXFR 2 Freeport, TX 206CAPR 3 Point Reyes Peninsula, CA 205CASD 4 San Diego, CA 202CAMD 5 Moss Landing, CA 197CAMR 5 Morro Bay, CA 197CAOC 5 Orange County (coastal), CA 197CACS 6 Crystal Springs, CA 194CASC 7 Santa Cruz County, CA 191TXSB 8 San Bernard N.W.R., TX 190TXBP 9 Bolivar Peninsula, TX 180CAOV 10 Oceanside-Vista-Carlsbad, CA 179CAAR 11 Arcata, CA 177CAHF 11 Hayward-Fremont, CA 177TXCC 12 Corpus Christi, TX 175CAMC 12 Marin County (southern), CA 175CARS 13 Rancho Santa Fe, CA 173CAWS 13 Western Sonoma County, CA 173CAMP 14 Monterey Peninsula, CA 172CAPP 14 Palos Verdes Peninsula, CA 172CAOA 15 Oakland, CA 171CAVE 15 Ventura, CA 171LASA 15 Sabine N.W.R., LA 171CACB 16 Centerville Beach to King Salmon, CA 170CAPA 16 Palo Alto, CA 170CAON 17 Orange County (northeastern), CA 169CASF 18 San Francisco, CA 167NJCM 18 Cape May, NJ 167NCSB 19 Southport, Bald Head & Oak Islands, NC 166TXAR 20 Aransas N.W.R., TX 165FLJA 21 Jacksonville, FL 164GASV 21 Savannah, GA-SC 164CASJ 22 San Jose, CA 163CABE 23 Benicia, CA 162TXPA 23 Port Aransas, TX 162CALU 24 La Purisima, CA 161TXGA 24 Galveston, TX 161NCWI 25 Wilmington, NC 160TXHO 25 Houston, TX 160FLNR 26 West Pasco (New Port Richey), FL 159CAMU 27 Malibu, CA 158FLNP 27 North Pinellas, FL 158LALT 27 Lacassine N.W.R.-Thornwell, LA 158TXAP 27 Attwater Prairie Chicken N.W.R., TX 158NCMC 28 Morehead City, NC 156VACC 28 Cape Charles, VA 156CALB 29 Long Beach-El Dorado, CA 155CASS 29 Salton Sea (south), CA 155CASU 30 San Juan Capistrano, CA 154FLAB 30 Aripeka-Bayport, FL 154FLCO 30 Cocoa, FL 154LACW 30 Crowley, LA 154ORCB 30 Coos Bay, OR 154CAAN 31 Año Nuevo, CA 153CAPC 31 Putah Creek, CA 153CASM 31 Sacramento, CA 153FLZE 31 Zellwood-Mount Dora, FL 153LACR 31 Creole, LA 153BCLA 32 Ladner, BC 152CASG 32 Santa Maria-Guadalupe, CA 152FLAL 32 Alafia Banks, FL 152FLMI 32 Merritt Island N.W.R., FL 152SCLP 32 Litchfield-Pawleys Island, SC 152TXCF 32 Corpus Christi (Flour Bluff), TX 152TXSA 32 Santa Ana N.W.R., TX 152SCHH 33 Hilton Head Island, SC 151CACC 34 Contra Costa County, CA 150LANO 34 New Orleans, LA 150MDOC 34 Ocean City, MD 150MSSH 34 Southern Hancock County, MS 150TXRO 34 Rockport, TX 150

Page 5: American Birds - National Audubon Society | Audubon

change (a.k.a. “global warming”), if theyresult in milder fall/winter seasons (anddespite the many negative aspects of suchevents), may tend to reduce the historicalspecies total advantage that counters insouthern latitudes have enjoyed.

Farther to the south, in Latin America,this season’s effort at Monteverde, CostaRica, tallied an astonishing 376 species oncount day, potentially the highest ever onany Christmas Bird Count. La Selva, alsoin Costa Rica, and Mindo-Tandayapa,Ecuador, also had species totals over 350.Can the elusive 400-species mark be faroff? The diversity of species in the neotrop-

ics is mind-boggling, and with increasingparticipation and coverage on counts inLatin America, a 400-species count daytotal seems possible.

Compilers and regional editors may bethought of as the heart and soul of theChristmas Bird Count, but its lifeblood isthe pool of participants. Of course, itshouldn’t be forgotten that most compilersand regional editors are included in the“observer” category as well. The ChristmasBird Count is somewhat unique in thatthere are two different types of census dataincluded: field observers and feeder watch-ers, with efforts for each kept separately.

There is no “good” or “bad” between thetwo, they are simply different types of cen-sus methods. Each is valuable in its ownright, and the combination serves to pro-duce an even better picture of the diversityand number of birds present in CBC cir-cles during the official count period ofDecember 14 through January 5 each sea-son. Table 4 this season is the list of allChristmas Bird Counts with 100 or moretotal observers reported. North Bay,Ontario, maintains its grip on the top spotin this category, a title belonging to thosededicated observers since the 99th CBC.

At this point in each of my summariesfor the past 14 years, it has been my tradi-tion to analyze four or five species’presence (or absence) in the given season’sChristmas Bird Count results. We’vealways taken an early-winter snapshot ofBohemian Waxwing and GrasshopperSparrow, often Evening Grosbeak andsome other “winter finch” or two, andoccasionally other species of note for thegiven year. This season I break with thattradition, and instead invite you to doyour own analysis. We have been workingvery hard over the past five years to bringthe Christmas Bird Count into the 21stcentury; the transition is now nearly com-plete. The data from each count everyseason are now entered on-line by com-pilers or their designated lieutenant, vastlyimproving the accuracy of data entry overthe old method of paper form submittalfollowed with data entry by contractors.The CBC historical database is availableon-line as well, and although it has beensince the 99th count, this year we’vedeveloped a new, interactive “output tool”that allows anyone to ask nearly any ques-tion of their own design of the CBCdatabase. Are you interested in the resultsof your count over its entire history?That’s available at the click of a mouse.Do you have an interest in a particularspecies on your count, in your region, oron the Christmas Bird Count as a whole?Just use that mouse again. Have youalways wanted to download the data fromyour count (or any other) into your owncomputer and do your own analyses? Goahead, visit the CBC home page,

THE 102ND CHRISTMAS BIRD COUNT AMERICAN BIRDS 5

Observers returning from a boat trip in Boston Harbor on the Greater Boston,Massachusetts, CBC noticed a large dark shape near the “Black Falcon Terminal” wherecruise ships dock. It wasn’t one of the city’s Peregrines, but instead a returning adultfemale Gyrfalcon (Falco rusticolus) that had been banded by Snowy Owl researcherNorm Smith at Logan airport in Boston some years before. This powerful predatorstayed in the city for much of the winter, to the delight of probably thousands of humanobservers; it was often observed taking gulls on the wing. Photo/Shawn P. Carey

Table 2b: Counts south of the United States–Mexican border

Count Code Rank Count Name Species Recorded

CRMO 1 Monteverde, Costa Rica 376

CRLS 2 La Selva, Lower Braulio Carillo, N.P., Costa Rica 365

ECNM 3 Mindo-Tandayapa, Ecuador 361

RPAC 4 Atlantic Canal Area, R.P., Panama 334

RPPC 5 Pacific Canal Area, R.P., Panama 312

RPCC 6 Central Canal Area, R.P., Panama 291

BLPG 7 Punta Gorda, Belize 267

BLBE 8 Belmopan, Belize 258

BLBC 9 Belize City, Belize 257

BLGJ 10 Gallon Jug, Belize 248

RPVC 11 Volcan, Chiriqui, Panama 219

MXES 12 Ensenada, Baja California, Mexico 189

TRTR 13 Trinidad, West Indies 172

MXYS 14 Yecora, Sonora, Mexico 152

BRIT 15 Itirapina, Sao Paulo, Brazil 151

ECCB 16 Cerro Blanco-Chognon-Puerto Hondo Estuary, Ecuador 150

Page 6: American Birds - National Audubon Society | Audubon

www.audubon.org/bird/cbc, and you’ll beable to do that, and many other things, aswell. And the current tool we have on theChristmas Bird Count home page is onlythe first generation of what we’re planningas a suite of tools that will allow anyone,with any need or interest, to makeinquiries of the entire CBC database.

You’ve undoubtedly noticed that this102nd Christmas Bird Count issue ofAmerican Birds is slimmed down fromany CBC volume that has been sent outin the past 50 years or so. What are notincluded in print this season are thecomplete results—every single bird andobserver—reported from every individ-ual Christmas Bird Count. Those dataare completely, and even more interac-tively, available on-line through theCBC home page.

In recent years the Christmas BirdCount issue has only been sent to com-pilers, regional editors, and a very fewsubscribers. This year we have decidedto expand readership, and have sent thebird count issue of American Birds tonearly all CBC participants. This vol-ume includes five feature articles, allhighlighting the value of the data thateach and every Christmas Bird Countobserver, compiler, and regional editorhas collected and reviewed over the past102 years. Two features are reprints ofpeer-reviewed scientific papers, one is adiscussion of the methods of analyses ofthe CBC database, another documentsthe value of the Christmas Bird Countfor monitoring exotic and introducedspecies, and the fifth highlightsAudubon WatchList bird species asreflected in the CBC historical data-base. These last two features includegraphs that were generated using theon-line output tool—the exact sametool that is now available for your per-sonal inquiries. At long last, themajority of participants on theChristmas Bird Count are able to seethe value of their work—the greatimportance of the CBC database toboth pure ornithology and to conserva-tion science. All of us are making adifference with our efforts on the count.

6 AMERICAN BIRDS

Table 3. Regional high counts for the 102nd (2001-2002) Christmas Bird CountRegion # of CBCs Highest Count (species total)St. Pierre & Miquelon 2 Ile St. Pierre (49)Newfoundland 13 St. John’s (62)Nova Scotia 18 Halifax-Dartmouth (116)Prince Edward Island 3 Prince Edward Island N.P. (62)New Brunswick 5 Sackville (60)Quebec 27 Quebec (85)Ontario 102 Long Point (119)Manitoba 17 Winnipeg (57)Saskatchewan 12 Fort Walsh, Cypress Hills (51)Alberta 36 Calgary (69)British Columbia 64 Ladner (152)Nunavut 2 Arviat (1)

Rankin Inlet (1)Northwest Territories 3 Norman Wells (16)Yukon Territory 5 Whitehorse (24)

Alaska 34 Juneau (75)Maine 25 Greater Portland (103)New Hampshire 16 Coastal New Hampshire (114)Vermont 16 Ferrisburg (85)Massachusetts 33 Cape Cod (133)Rhode Island 3 Newport County-Westport (124)

South Kingstown (124)Connecticut 16 New Haven (126)New York 72 L.I.: Sagaponack (137)New Jersey 27 Cape May (167)Pennsylvania 66 Southern Lancaster County (105)Delaware 7 Bombay Hook N.W.R. (135)

Rehoboth (135)Maryland 23 Ocean City (150)District of Columbia 1 Washington (115)Virginia 40 Cape Charles (156)North Carolina 44 Southport, Bald Head, & Oak Islands (166)South Carolina 18 Litchfield-Pawleys Island (152)Georgia 21 Savannah (164)Florida 61 Jacksonville (164)Ohio 54 Cincinnati (88)

Toledo (88)West Virginia 16 Charles Town (80)Kentucky 10 Land Between the Lakes (91)Tennessee 25 Reelfoot Lake (114)Alabama 12 Gulf Shores (144)Mississippi 15 Southern Hancock County (150)Michigan 53 Anchor Bay (86)Indiana 38 Lake Monroe (106)Wisconsin 42 Madison (89)Illinois 54 Rend Lake (104)Minnesota 47 Duluth (74)Iowa 28 Keokuk (87)Missouri 25 Mingo (105)Arkansas 21 Holla Bend N.W.R. (112)Louisiana 22 Sabine N.W.R. (171)North Dakota 19 Garrison Dam (57)South Dakota 16 Pierre (69)Nebraska 9 Lake McConaughy (103)Kansas 24 Linn County (94)Oklahoma 20 Oklahoma City (108)Texas 95 Mad Island Marsh (233)Montana 33 Bigfork (82)

Stevensville (82)Wyoming 18 Casper (64)Colorado 37 Pueblo Reservoir (110)New Mexico 30 Caballo (131)Idaho 22 Hagerman Valley (92)Utah 16 Provo (101)Nevada 11 Truckee Meadows (110)Arizona 33 Green Valley-Madera Canyon (147)Washington 42 Sequim-Dungeness (140)Oregon 41 Coos Bay (154)California 114 Santa Barbara (206)Hawaii 11 Honolulu, O’ahu (52)Palmyra Atoll 1 Palmyra Atoll (14)Guam 2 Dededo (29)Saipan 2 Saipan (41)Mexico 12 Ensenada, Baja California (189)Belize 4 Punta Gorda (267)Costa Rica 2 Monteverde (376)Panama 4 Pacific Canal Area (334)Colombia 3 Sabana de Bogota (140)Ecuador 2 Mindo-Tandayapa (361)Peru 1 Rio Orosa, Loreto (129)South Atlantic Ocean 1 Drake Passage (23)Brazil 2 Itirapina, Sao Paulo (151)Paraguay 1 Asuncion Bay, Asuncion (102)Trinidad 1 Trinidad (172)Bahamas 2 New Providence Island (114)Dominican Republic 2 Santo Domingo (50)Puerto Rico 3 Cabo Rojo (111)Virgin Islands 3 St. Croix (65)

Tortola (65)Bermuda 1 Bermuda (89)

Page 7: American Birds - National Audubon Society | Audubon

The future is very bright for theChristmas Bird Count on several differ-ent fronts. Based upon the highlysuccessful model of Audubon’s partner-ship with Bird Studies Canada as theCanadian partner in the count, we hopeto engage other organizations in LatinAmerica and the Caribbean to facilitatethe growth of the count in areas outsidethe United States and Canada. A “scien-tific review” of the count will beundertaken (as with the Breeding BirdSurvey) to discuss the pros and cons ofcurrent CBC methodology and todetermine ways of enhancing the valueof the Christmas Bird Count data toresearchers interested in studying earlywinter (that’s winter in North America)bird populations across the WesternHemisphere. And finally, in addition todeveloping new output tools for the his-torical CBC database, we’ve begun theprocess of revamping the compiler’sdata entry pages. This process will resultin an easier, faster, and generally more“user friendly” data entry experience forChristmas Bird Count compilers.

Since the horrific events ofSeptember 11, 2001, the perception ofthe world has changed for many peoplein North America. The effect of drasti-cally restricted access to U.S. militaryand government facilities was directlyfelt by many compilers during the102nd Christmas Bird Count, asheightened security precautions nega-tively affected the organization of manyCBCs that are conducted in such loca-tions across the continent. Many areaswere “off limits” to the public, or onlyavailable for CBC census to a limitedobserver pool, or under official escort. Afew counts had to be cancelled com-pletely. But the vast majority of existingChristmas Bird Counts were conduct-ed, many new counts were started, anda record number of observers turnedout to be part of the 102nd Count.Humans strive to find solace during dif-ficult times, and “getting back tonature” can provide a soothing effect forpeople who appreciate the naturalworld. The Christmas Bird Count has a

great lore and tradition that is part andparcel of the program, and it has beenconducted without interruptionthrough five prior wars. From time totime there has been a diminution in thenumber of observers and counts due tosuch tragedies, just as there has beenduring seasons of difficult treatmentfrom Mother Nature.

Finding solace in nature during thewinter solstice may be a part of the tradi-tion for many people involved with theChristmas Bird Count, no matter whatthe catalyst that feeds their need. It’s beenthis way for over a century, and is likely tocontinue for many decades to come.Perhaps the most amazing fact is that thedata that are obtained by all the dedicat-ed CBC observers serve to help ensurethe survival of the very creatures thatdrive the passions of Christmas BirdCount participants. My own associationwith the Christmas Bird Count beganduring graduate school about a quarter ofa century ago, when I first began partici-pating on counts; I could never haveimagined the twists of fate that wouldlead to my being in charge of the entireprogram. Some of the most wonderfulbirders, ornithologists, conservationists,and human beings I’ve ever encounteredhave been involved with the ChristmasBird Count. We salute the dedicatedmultitudes of CBC observers each sea-son, and anticipate meeting you in thefield (or at feeders) during futureChristmas Bird Count seasons.

THE 102ND CHRISTMAS BIRD COUNT AMERICAN BIRDS 7

Table 4.Counts with 100 or more participants on the102nd (2001-2002) Christmas Bird Count

(fw = feederwatchers)

Count Count Name # ParticipantsCode

ONNB North Bay, ON 889 (44 + 845 fw)ABED Edmonton, AB 561 (196 + 365 fw)CAWB Wallace-Bellota, CA 527 (37 + 490 fw)MACO Concord, MA 269 (87 + 82 fw)OHTO Toledo, OH 267 (31 + 236 fw)BCVI Victoria, BC 197 (193 + 4 fw)CAPR Point Reyes 184 (184 + 0 fw)

Peninsula, CACAOA Oakland, CA 177 (163 + 14 fw)CASB Santa Barbara, CA 176 (172 + 4 fw)VAFB Fort Belvoir, VA 172 (170 + 2 fw)ABCA Calgary, AB 171 (70 + 101 fw)CTHA Hartford, CT 170 (120 + 50 fw)ABSA St. Albert, Alberta 164 (74 + 90 fw)BCVA Vancouver, BC 150 (116 + 34 fw)SCHH Hilton Head Island, SC 149 (149 + 0 fw)MBWI Winnipeg, MB 147 (80 + 67 fw)OHCF Cuyahoga Falls, OH 144 (88 + 56 fw)ABSR Strathcona, AB 140 (49 + 91 fw)BCNN Nanaimo, BC 139 (108 + 31 fw)NSWO Wolfeville, NS 139 (48 + 91 fw)LABR Baton Rouge, LA 136 (59 + 77 fw)AKFA Fairbanks, AK 135 (100 + 35 fw)SKSA Saskatoon, SK 131 (77 + 54 fw)MIPO Pontiac, MI 129 (73 + 56 fw)NSHD Halifax-Dartmouth, NS 129 (77 + 52 fw)AKAN Anchorage, AK 127 (80 + 47 fw)RIBI Block Island, RI 127 (127 + 0 fw)CAMC Marin County 126 (120 + 6 fw)

(southern), CAOREU Eugene, OR 126 (99 + 27 fw)SCSC Sun City-Okatie, SC 126 (118 + 8 fw)WASE Seattle, WA 126 (123 + 3 fw)NYIT Ithaca, NY 123 (91 + 32 fw)FLSC Sanibel-Captiva, FL 122 (120 + 2 fw)CASF San Francisco, CA 120 (110 + 10 fw)SASJ San Jose, CA 120 (120 + 0 fw)PAPI Pittsburgh, PA 119 (88 + 31 fw)TXMM Mad Island Marsh, TX 119 (115 + 4 fw)COBO Boulder, CO 116 (87 + 29 fw)DCDC Washington, DC 115 (114 + 1 fw)ONOH Ottawa-Hull, ON 113 (99 + 14 fw)MANO Northampton, MA 112 (94 + 18 fw)CASD San Diego, CA 110 (110 + 0 fw)BCWR White Rock, BC 108 (91 + 17 fw)WASD Sequim-Dungeness,WA 108 (80 + 28 fw)CAOC Orange County 106 (105 + 1 fw)

(coastal), CACODE Denver, CO 106 (65 + 41 fw)TXFR Freeport, TX 106 (102 + 4 fw)BCKE Kelowna, BC 105 (62 + 43 fw)PACH Chambersburg, PA 105 (80 + 25 fw)ILFB Fermilab-Batavia, IL 104 (102 + 2 fw)QCQU Quebec, QC 104 (104 + 0 fw)MDSE Seneca, MD 101 (100 + 1 fw)ONLO London, ON 100 (56 + 44 fw)

Page 8: American Birds - National Audubon Society | Audubon

As the Christmas Bird Count seasonopened in mid-December 2001, there wasan air of expectancy across Canada thathadn’t been as strong for years. Most areashadn’t had a significant cold snap, water-courses were at least partly ice-free, andmany sites across the southern part of thecountry didn’t have any amount of snow.One of the first birds I saw on thePenticton, British Columbia, count was aHermit Thrush, the first record in thecount’s 44-year history. A few days later ourteam counted not one but two CommonYellowthroats on the Oliver-Osoyooscounts, another new species for that long-standing count. And there were stories likethat across the country, many of themmuch more surprising, such as the Black-necked Stilt at Wallaceburg, Ontario, andthe Blue-winged Warbler at Halifax-Dartmouth, Nova Scotia.

This is the second year of ChristmasBird Count coordination for Bird StudiesCanada, and the two years could hardly bemore different. This year, Canadian CBCparticipants reported 3.7 million individualbirds, up more than 25 percent from lastyear’s 2.9 million. The species tally was upas well, from 283 to 299 species. As theQuebec CBC regional editor, MarcelDarveau, points out, it may not be just the

warm weather allowing more species towinter at the northern end of theirranges—pleasant weather also encouragesmore observers to go out counting, andstay out longer on count day.

Whether it was the warm days, the clearroads, or simply the increasing popularityof birding, the number of field observers onCanadian Christmas Bird Counts was upmore than ten percent this year to 7221.The count total rose to 310, compared to298 last year; of all the new counts we espe-cially welcome Arviat and Rankin Inlet,giving us coverage in Nunavut. The fullprovincial and territorial breakdown isshown in Table 1. North Bay again had thehighest total of participants, with 889, ofwhich 845 were feeder watchers. BothEdmonton and Victoria upped their fieldobserver numbers substantially, butEdmonton held onto that crown with 196,while Victoria had 193. The other countswith more than 100 field observers wereVancouver, Nanaimo, and Quebec.

Local rivalries in Christmas Bird Countsinvariably focus on species totals, and thisyear top bragging rights goes to Ladner, thecount that covers the rich Fraser River deltain southwestern British Columbia. Oftenplagued with rain, fog, or wind, Ladnerhad a perfect day this year and some verygood birds (including an Ivory Gull) topost 152 species, tying the all-time recordfor species total held by Victoria. LongPoint broke its old Ontario record with119 species. While most other counts inCanada can’t hope to ever break 100species, let alone 150, local records fell likedominoes across the country this year. Onerecord of note was Fort Walsh’s total of 51species, the first count to ever break 50 inSaskatchewan. Winnipeg bested by fivetheir own record for a Manitoba count (setonly last year) with 57 species.

Ladner again topped the individualtotals with 250,594 birds, mostly Dunlin,American Wigeons, Glaucous-wingedGulls, Mallards and Snow Geese. Othermembers of the “Hundred ThousandClub” this year were Prince Edward Point(162,454, almost entirely Long-tailedDucks) and Blenheim (121,574, mostlyscaup and Ring-billed Gulls).

It is interesting to look at the changesin the counts of the most common species

8 AMERICAN BIRDS

Christmas Bird CountCanada: 2001-2002

Richard J. Cannings

Table 1. Provincial and territorial summaries for the 102nd Christmas Bird Count.

Province/ Counts Individuals Species Field Observers Feeder-watchers Highest speciesTerritory total and countBC 64 1,121,580 231 1902 497 152, Ladner

AB 36 195,954 118 886 988 69, Calgary

SK 12 47,340 80 154 93 51, Fort Walsh

MB 17 65,730 81 273 138 57, Winnipeg

ON 102 1,735,834 188 2572 1807 119, Long Point

QC 27 200,220 135 573 103 85, Quebec

NB 5 20,573 90 96 9 60, Sackville

PEI 3 46,888 89 56 2 62, PEI National Park

NS 18 221,272 165 419 324 116, Halifax-Dartmouth

NF 14 52,794 108 167 67 62, St. John’s

YT 5 4055 31 51 16 24, Whitehorse

NWT 3 2157 24 40 2 16, Norman Wells

NU 2 16 1 3 0 1, Arviat, Rankin Inlet

SPM 2 6595 63 29 0 49, Île St. Pierre

TOTAL 310 3,721,008 299 7221 4046

Page 9: American Birds - National Audubon Society | Audubon

THE 102ND CHRISTMAS BIRD COUNT AMERICAN BIRDS 9

(Table 2). All the waterfowl from central andeastern Canada were up in numbers, partic-ularly Long-tailed Duck (although almost80 percent of the country’s Long-tailedswere on one count—Prince Edward Point,ON). Ring-billed Gulls were also wide-spread and abundant; Point Pelee, Ontariohad almost as many Ring-billeds as wereseen across the entire country last year. Onthe west coast, numbers of the largest ofCanadian waterfowl, the Trumpeter Swan,increased once again to 6964.

Some of the northern finches movedsouth in impressive numbers this year. PineGrosbeak reports more than doubled, withover 22,000 seen. But it was the year of theredpoll, especially in western and centralCanada. Last year a grand total of 5835Common Redpolls were reported inCanada, this year that total soared to96,395. Numbers of Pine Siskins, a closecousin of the redpolls, doubled this year toover 70,000. Unlike the highly skewedLong-tailed Duck reports, redpolls weredistributed rather evenly across the countryand in fact were tied for third in list of birdsseen on the most counts (Table 3).

Since most counts are located near thesouthern edge of Canada, the true test of awidespread species is whether it was seen inevery province and territory, and this yearonly one species made that list—CommonRaven. Birders at Arviat and Rankin Inlet,the only counts in Nunavut, could only

find one species on count day—raven. There were so many reports of interest-

ing birds this year it’s difficult to knowwhere to begin. On the Atlantic coast anunprecedented array of warblers and otherspecies not usually associated withCanadian winters hung on after a sum-mer-like fall into the fall-like winter. FourAsh-throated Flycatchers were scatteredthrough the Maritimes last December, butonly one was found on count day—atSackville, New Brunswick—only the sec-ond ever for a Canadian CBC. AWhite-eyed Vireo stuck in the stuntedboreal forests of Cape St. Mary’s,Newfoundland was completely out ofplace. Nova Scotia birders found 12species of warblers, eight of them atHalifax-Dartmouth alone.

Reports of northern owls were downsignificantly this year after last year’s flightof Northern Hawk and Great Gray Owls,but Snowy Owls were much commonerthan usual in Atlantic Canada and the BCInterior. Alberta owlers are definitely thekeenest in the country, and this year theirefforts truly paid off with outstandingnumbers for several species, includingBoreal and Northern Saw-whet.

While it sounds like field observers hadan easy time of it this year, we are still talk-ing about winter in Canada, and I’d liketo personally thank all the observers whoparticipated in the Christmas Bird Count.

One observer in particular deserves specialpraise this year—Doug Brown ofOsoyoos, British Columbia. While Doughas always done multiple counts eachChristmas, this year he outdid himself,taking part in 17 counts, driving 6700kilometres and seeing 150 species whiledoing so! I’d also like to thank all thecount compilers for their hard workorganizing participants and entering thedata on the website. Without you therewouldn’t be a Christmas Bird Count.

Table 2. The top 15 bird species by total number on the 102nd Christmas Bird Count inCanada, with a comparison of results from the 101st count.

Species Rank in Rank in Total in Total in Percent 101st 102nd 101st 102nd changeCount Count Count Count

European Starling 1 1 327,825 412,908 26.0

American Crow 2 2 247,369 246,627 -4.3

Canada Goose 5 3 119,675 236,760 97.8

Mallard 3 4 161,270 220,214 36.5

Long-tailed Duck 12 5 71,020 186,460 162.5

House Sparrow 4 6 122,300 142,345 16.4

Rock Dove 6 7 98,815 137,259 38.9

Ring-billed Gull 33 8 21,784 128,703 490.8

Black-capped Chickadee 7 9 98,248 123,571 25.8

Herring Gull 10 10 80,443 117,855 46.5

Glaucous-winged Gull 8 11 91,092 100,746 10.6

Common Redpoll 62 12 5835 96,395 1552.0

Dunlin 9 13 84,728 71,852 -15.2

Pine Siskin 23 14 34,417 70,531 104.9

Greater Scaup 18 15 38,527 64,580 67.6

Table 3. The most widespread species onCanadian Christmas Counts, as measuredby the number of counts on which theywere reported.

Rank Species Numberof counts

1 Black-capped Chickadee 280

2 Dark-eyed Junco 278

3 Hairy Woodpecker 273

4 Downy Woodpecker 273

5 Common Redpoll 273

6 Red-breasted Nuthatch 272

7 European Starling 268

8 Common Raven 255

9 House Sparrow 253

10 Rock Dove 245

11 Pine Grosbeak 235

12 Blue Jay 224

13 American Crow 211

14 Mallard 207

15 White-breasted Nuthatch 205

Warm fall and early winter weather allowedmore southern birds to survive farther norththis year, including this Summer Tanager(Piranga rubra) in a bizarre setting at DeepRiver, Ontario. Photo/Chris Michener

Page 10: American Birds - National Audubon Society | Audubon

The Christmas Bird Count (CBC) is aunique historical database on bird popu-lations, and in recent years the ornitho-logical community has been placing moreand more demands on the information itcontains. Natural historians, conserva-tionists, wildlife managers, and scientificornithologists use the CBC to makemaps of winter bird distributions to eval-uate how these distributions are influ-enced by environmental features and cli-mate change (e.g., Root 1988), and toestimate population change (Sauer et al.1996, Dunn and Sauer 1997).

The scientific credibility of theseresults depends on the development anduse of appropriate methods of statisticalanalysis that identify and accommodatethe limitations of the survey design.Unfortunately, statistical methodsapplied in analysis of CBC data are oftenoverly simple, and lead to flawed analysis.For example, comparisons of simple aver-ages of CBC data are often misleading asmeasures of population change over spaceor time. To make reasonable inferencesabout bird populations from CBC data,it is usually necessary to use statisticalmethods specifically developed for thesurvey. Unfortunately, these methods arenot well known to many scientists.

To the casual observer, the running ofa CBC looks quite chaotic, with bothskilled and unskilled observers countingportions of each circle, methods of count-ing varying among circles, and sometimesuse of specialized counting methods fordifficult-to-identify or rare taxa such asgulls and eagles. Abundant data are col-lected near cities but few circles exist inremote regions. Those of us who activelyparticipate in CBCs may occasionallyconsider how complicated the relation-ship must be between the number ofbirds counted during a CBC and theactual population size in the circle. Also,since the first CBC in 1900, both thenumber of observers and the number of

circles have increased greatly. Obviously,if more circles are counted and moreobservers participate, more birds arecounted; changes in counts reflect thesechanges in the survey. How do we extractthe “signal” (change in bird populations)from noise in the data due to thesemethodological problems?

Design of Wildlife Surveys

The CBC is quite typical of many his-torical datasets. It was developed for rea-sons vastly different from many of itsmodern uses, and it seems a bit unfair tocriticize the survey design based on mod-ern goals and standards. However, theonly way to avoid flawed analyses is todetermine whether the CBC meets mod-ern standards, and if it does not, to deter-mine whether modern analysis methodscan at least partially accommodate thedesign limitations of the survey. Two crit-ical components of survey design are: (1)selecting sample locations so that thesample is spatially representative of thepopulation of interest, and (2) under-standing how the counts collected in thesurvey are related to the actual popula-tion sizes at the sampled sites. The scien-tific field of survey sampling has empha-sized (1), but (2) has been a topic of par-ticular interest to wildlife statisticians.When surveys do not meet acceptedstandards with regard to either of theseattributes, one can still analyze theresults, but inferences are weaker as wemust make assumptions (that are usuallyuntestable) about how the survey datarelate to the actual bird populations.

Choice of Sample Locations

One goal of survey design is to ensurethat sample sites are selected in such away as to allow inference to a larger pop-ulation. This can be accomplished by avariety of random sampling schemes,collectively known as probability sam-pling. Sites that are easily sampled or are

located in places where people are inter-ested in counting, such as nationalwildlife refuges or national parks, arelikely not to be representative of largerareas. An enormous statistical literatureexists on how to select samples for infer-ence, and methods such as stratification,dual frame sampling, and adaptive sam-pling all provide efficient means ofselecting samples that still preservesome notion of probability sampling(Thompson 1992).

Unfortunately, most CBC circles arenot selected in conjunction with a statis-tical design, but reflect sample locationsaccessible to counters or of particularinterest to local coordinators. The ten-dency to place circles near areas of par-ticular bird abundance ensures that birdswill be available to count, but also virtu-ally guarantees that the sampled area isnot representative of the surroundingregion. A cautious approach to use ofCBC data would require that inferencebe restricted to the sites that are actuallysampled. Results summarized for a largergeographic area from the survey shouldhave a caveat label “from CBC circleswithin” to ensure that users understandthe lack of representativeness.

There are cases in which the CBC sam-ple may be used for reasonable inference,even though the sites are not representa-tive of the landscape. First, often the areacovered by the circle (or a collection ofcircles) is itself of primary interest, andthe circle data can be used in evaluationsof population change. Second, eventhough the sites themselves are not ran-domly located, it is sometimes possible toextrapolate to the region of interest bymaking assumptions about how theinformation at a site relates to the region-al information, then using the modeldefined by these assumptions to extrapo-late to the region of interest. This model-based approach is generally used in anyregional analysis of CBC data. For

10 AMERICAN BIRDS

Using Christmas Bird Count Data in Analysis of Population ChangeJohn R. Sauer and William A. Link

USGS Patuxent Wildlife Research Center, 11510 American Holly Drive, Laurel, MD 20708-4017

Page 11: American Birds - National Audubon Society | Audubon

THE 102ND CHRISTMAS BIRD COUNT AMERICAN BIRDS 11

example, by defining strata for analysis(e.g., physiographic regions within states),the sample is partitioned into strata with-in which the circles are at similar densities,hence accommodating regional patternsin sample intensity. However, any model-based approach to analyses of the surveydoes impose a series of assumptions thatmust be stated explicitly. For example, inthe stratum-based approach, there is anassumption that within the region theCBC circles are a reasonable sample. Thisassumption is untested, but unlikely to betrue. Because all reasonable analyses ofCBC data are model-based, it is clear thatthe success of the analysis depends ondevelopment of clearly defined and rea-sonable assumptions.

Relating Counts to Population Sizes

A second controversial aspect of sam-pling from the CBC is the issue ofdetectability of birds within sites. Almostno bird species are truly censused (i.e.,completely counted) within CBC circles,and no direct method exists to estimatedetectability of birds in CBCs. In orderto conduct any analysis of CBC data wehave to make some assumptions aboutthe relationship between the counts ofbirds within circles and the actual popu-lation sizes. This necessity is commonwith bird surveys. Counts that are notclearly related to the actual populationsize are called indices. Unfortunately,any analysis of indices requires anassumption of consistency: If indices areto be used as surrogates for populationsizes in analysis of population change, itmust be assumed that patterns indetectability are not confounded withpatterns in population size.

Detectability problems bedevil all birdsurveys, and even the most stubborn fieldbiologist can recognize that somethinghas to be done to accommodatedetectability differences in counts fromthe CBC. A plethora of methods havebeen proposed for estimating detectabili-ty during field studies of birds, and someof these could (but have not as yet) beapplied during CBC sampling. However,we must do something to accommodate

differences in detectability over time andspace in the historical data. A commonstrategy is to identify some observablefactor that could be associated with varia-tion in detectability, to model this varia-tion, and ultimately, to adjust the countsfor variation in detectability. For theChristmas Bird Count, the most obviousfactor associated with counting is effortexpended in the circle. Effort is recordedfor a variety of activities associated withCBCs, and total party-hours is mostoften used as a covariate representingeffort in counting.

Adjusting Counts for Effort

The effects of effort on the countsobtained in CBC circles are undoubtedlyenormously complicated. No one ques-tions that, for the bulk of CBC species,there is some association between effortand counts. However, it is not a simpleassociation and could be influenced bymany factors. For example, some speciesare frequently only counted by a speciallydesignated skilled observer, rare speciesand unique habitats are often coveredregardless of effort, and the differenttypes of effort could have very differentassociations with counts. Clearly, somethought is necessary before attempting toapply effort adjustments to any counts,and the form of the effort adjustment islikely to vary among species. Nevertheless,changes in effort have been so great overthe interval of the CBC (see Butcher andMcCulloch 1990 for a summary ofchanges in effort over time), that failureto consider effort as a factor influencingcounts will undermine the credibility ofany analysis of CBC data. A flexibleapproach to modeling the influence ofeffort on counts is needed.

Historically (Bock and Root 1981,Dunn and Sauer 1997), analyses havesimply divided counts by party-hours orparty-miles. (Here, we do not dissect thevarious kinds of effort, but simply con-sider total party-hours as the effortindex). This simple scaling has some lim-itations, and in particular it does notallow for the possibility that increasedeffort may have less effect when effort is

high. Instead, additional effort is alwaysassumed to influence counts regardless ofthe amount of effort already occurring.Butcher and McCulloch (1990) suggest-ed a nonlinear adjustment, and Link andSauer (1999a,b) proposed a general set ofeffort adjustments similar to the Box-Coxfamily of transformations that permit avariety of alternative relationshipsbetween counts and effort. This effortrelationship has form (ξp -1)/p, where ξ isthe effort value and p is an exponent thatchanges the shape of the relationship ofeffort to count.

This family of curves includes a varietyof realistic forms. Some allow the effect ofeffort to have an asymptote, to allowdiminishing returns; others include thesimple effort adjustment (as p approaches0) and the Butcher and McCulloch(1990) adjustment (as p approaches 1). Toactually find the appropriate effort adjust-ment, Link and Sauer (1999a,b) suggestconducting the analysis of interest (e.g.,fitting a linear model to the data) using avariety of values of p, then choosing thevalue of p that minimizes the scaleddeviance (a measure of goodness of fit).

An Example of This Analysis: Population

Change in Red-tailed Hawks

The Red-tailed Hawk (Buteojamaicensis) has a widespread winter dis-tribution in North America (Figure 1).We estimated population change forwintering North American Red-tailedHawks using a generalized linear modelthat included effort adjustments.Because the CBC sample has a great dealof spatial heterogeneity, we chose to ana-lyze the results by Bird ConservationRegions (BCRs). These regions weredeveloped to provide a common frame-work for bird conservation, and general-ly tend to divide the continent intoregions containing consistent habitats.We used CBC data for the period from1955 to 1999 in this analysis.

The actual model we fit was a “yeareffects” model, in which we estimatedseparate composite year indices for eachBCR using the log-linear model,

ln(µi,j) = θi + B (ξp -1)/p + ϕj

Page 12: American Birds - National Audubon Society | Audubon

where i indexes CBC circle, j indexesyear, µi,j is the expected count in the cir-cle, θi is a “baseline” abundance valuefor a circle, B is the slope parameterassociated with effort, and ϕj is the yeareffect corresponding to year j. This gen-eralized linear model was not based ona normal distribution, but instead per-mitted counts to be Poisson in distribu-tion and incorporated overdispersion inthe Poisson counts. Readers interestedin the technical details of fitting this loglinear model should read the discussionin Link and Sauer (1999b).

In our analysis, we fit the year effectsmodel for each stratum, using a series ofeight candidate exponents (p = -2.0, -1.5,-1, -0.5, 0, 0.5, 1.0, 1.5) for the effect ofeffort. We selected the value of p thatminimizes the total deviance.

Getting composite estimates ofchange and annual indices from theCBC is complicated by regional differ-ences in precision of regional estimates.Some BCRs have very precise estimatesof change, while others are very impre-cisely estimated. Empirical Bayes proce-dures accommodate the differences inprecision by using all regional indices toestimate a set of prior mean year effects.Revised BCR year effects are then cal-culated as a weighted average of theBCR year effect and the prior meanyear effect, with weights depending onthe relative precision of the BCR yeareffects. The empirical Bayes estimatesthus are “shrunken” toward the priormeans, with the amount of shrinkageassociated with the amount of informa-tion contained in the original estimates(Link and Sauer 1998). After we calcu-late the empirical Bayes estimates ofyear effects for each BCR, we then scalethem to reflect a comparable level ofeffort among regions, and calculate acomposite annual index as an area-weighted average of the BCR indices.Note that we use the term “year effect”to denote the relative measures of year-ly change from the generalized linearmodel analysis, but “annual indices” areyear effects that are scaled by a medianeffort for the region.

12 AMERICAN BIRDS

Figure 1. Map of winter distribution of Red-tailed Hawks from CBC data. These mapswere produced using a simple index of mean (1966–1989) abundance by circle(counts/total party hours), and smoothed using a distance-weighted average ofinformation from nearby CBC circles.

Figure 2. Population trajectories for Red-tailed Hawks in the Southeastern CoastalPlain Bird Conservation Region from CBC data, using three alternative exponents forthe relationship between effort and count.

Figure 3. Composite annual indices for Red-tailed Hawks from the CBC data,1955–1999, with associated 95 percent confidence intervals.

Page 13: American Birds - National Audubon Society | Audubon

THE 102ND CHRISTMAS BIRD COUNT AMERICAN BIRDS 13

Effects of Effort Adjustment

Results of the analysis for theSoutheastern Coastal Plain BCR clearlydocument the effects of choice of theexponent p on the analysis (Figure 2).Negative values of p tend to limit theinfluence of large effort values butincrease the effects of small changes ineffort when effort is low. On the otherhand, positive values of p tend to retainan increasing effect of effort even wheneffort is high. Consequently, the analysisfit with the highest value of p (1.5) sug-gests an increase of about 29 percentover the interval, while indices fit withthe lowest p (-2.0) indicate a populationincrease of 16 percent over the interval.It is clear that choice of p (and hence ofthe shape of the effort adjustment) canhave important consequences for theestimation of population change fromthe CBC.

Population Change Analysis

We used information from 2681 cir-cles and calculated annual indices forRed-tailed Hawks from 30 BCRs. Whenwe sum up the deviances from the analy-sis of the 30 BCRs, the minimumoccurred with p = -2.0. Consequently, weused this effort adjustment in summaryanalyses. The composite annual indices(Figure 3) indicate an increasing popula-tion over the interval 1955–1999. Thecomposite index value at 1955 (7.69, 95percent confidence interval: 6.62, 8.76) iswell below the index value in 1999(9.24), indicating a population increaseof 20.1 percent over the interval. Yeareffects and variances are all relative to thefinal year, hence the variance of the finalyear effect is 0.

The North American Breeding BirdSurvey (BBS, Robbins et al. 1986) hasextensive data on breeding populationsof Red-tailed Hawks for the regions sur-veyed by the CBC. Even though theactual breeding and wintering popula-tions are unlikely to match, it is of gen-eral interest to compare the patterns ofcontinent-scale population indices pro-duced by the surveys. To facilitate com-parison, we scaled both time series of

annual indices by subtracting the meanof the series then dividing by the vari-ance of the indices from each value(Figure 4). The general pattern ofincrease over time in the population isconsistent between the two sets ofresults, although patterns of year-to-yearchange tend not to coincide. Althoughdetailed discussions of the associationsbetween annual indices of wintering andbreeding populations is beyond thescope of this paper (see Dunn and Sauer[1997] or Sauer et al. [1998] for furtherresults and discussion of this topic), it isnot surprising to observe a lack of year-to-year coincidence; clear differencesexist in precision between the surveys, inpart associated with unmodeled effectsof winter weather in CBC tallies.

Accommodating Survey Design Issues

During the Analysis of CBC Data

The key to any successful analysis ofCBC data is to begin with a carefulreview of how the limitations of the dataare likely to influence the results of theanalysis, then to choose methods ofanalysis that accommodate as much aspossible the limitations of the survey.For example, in our analysis we includ-ed a flexible model for effort adjust-ment, then used information from thedata to guide the selection of the bestmodel. We then stratified the analysis byBCRs, which provided some geographicstructuring to accommodate the region-al variation in number of samples. We

used a model that allowed for overdis-persed Poisson data appropriate forcounts, and employed empirical Bayesprocedures to accommodate differencesin quality of information in regionalsummaries. The generalized linearmodel approach is very flexible, and canbe applied to a variety of studies focusedon factors influencing wintering birdpopulations. In particular, the modelcan be easily modified to contain covari-ates, allowing for assessment of associa-tions between CBC counts and winterweather, disturbance, and a variety ofother environmental factors.

All count-based surveys have the lim-itation that it is often difficult to distin-guish factors that influence counts fromfactors that influence the actual popula-tions (e.g., Bennetts et al. 1999). For theCBC, it is clear the effort influencescounts, and hence, that increases ineffort over time must be accommodatedin any analysis. Because it is difficult tospecify exactly how effort will influencecounts, methods such as we present hereare important in that they provide toolsfor quantitative assessment of the formand significance of the effort adjust-ment. Of course, the approaches toeffort adjustment are all model-based.We must assume that the modelsapplied to the data are actually appropri-ate; misspecification may introduce bias.Unfortunately, effort adjustment is likelyto be quite subtle, and variation amongregions could introduce heterogeneity

Figure 4. Annual indices for breeding populations and wintering populations ofRed-tailed Hawks in North America, based on data from the North American BreedingBird Survey and the CBC. Results are scaled to facilitate comparisons.

Page 14: American Birds - National Audubon Society | Audubon

that is difficult to model. Uncertaintywill always exist in modeling the associ-ation between effort and counts, andthis uncertainty limits our use of CBCdata. It would be useful to considermodifying the field procedures for datacollection so that changes in effort arenot associated with changes in counts,and to evaluate experimentally alterna-tive approaches to counting birds inCBCs that would allow estimation ofdetectability as part of the design (e.g.,through double counting methods,Nichols et al. 2000).

Analyses and Survey Methods Should Be

Subject to Constant Revision

The generalized linear modelapproach we describe is only one ofmany approaches that can be used toestimate population change from CBCcounts and associated effort informa-tion. Other procedures that could beused for CBC analyses include smooth-ing procedures such as generalized addi-tive models (e.g., Fewster et al. 2001).We have recently implemented a hierar-chical modeling approach for BBS datathat allows for summary among regionsas part of the primary analysis ratherthan as a separate procedure (Link andSauer 2002), and intend to apply similarmodels to CBC data, allowing for directestimation of composite annual indices.

These new methods for survey analy-sis provide insights into changes in sur-vey design that can enhance the value ofthe information. The CBC has beenextremely successful as a tool for increas-ing public interest in birding and birdconservation. Use of the information forbird conservation creates new demandson quality of information. It is impor-tant to maintain a dialogue betweenusers of the information, informationneeds for the analyses, and survey coor-dinators and participants. Our work assurvey analysts emphasizes the value andlimitations of existing data, and providessome indications of what features of thesurvey could be modified to make thesurvey a more reliable source of birdpopulation data. These results should

feed back into decisions on future surveymethods and design. For example, itwould be useful to further evaluatethe relationship between effort anddetectability of birds within circles, andto obtain better information regardingwithin-circle protocols. Surveys onlyremain useful if they adapt to currentneeds, while maintaining consistencywith historical goals.

Literature Cited

Bennetts, R. E., W. A. Link, J. R. Sauer,and P. W. Sikes, Jr. 1999. Factorsinfluencing counts in an annual sur-vey of snail kites in Florida. Auk116:316-323.

Bock, C. E., and T. L. Root. 1981. TheChristmas Bird Count and avian ecol-ogy. Pages 17-23 in C. J. Ralph and J.M. Scott, editors, Estimating numbersof terrestrial birds, Studies in AvianEcology 6.

Butcher, G. S., and C. E. McCulloch.1990. Influence of observer effort onthe number of individual birdsrecorded on Christmas Bird Counts.Pages 120-129 in J. R. Sauer and S.Droege, editors, Survey designs andstatistical methods for the estimationof avian population trends, U.S. Fishand Wildlife Service, BiologicalReport 90 (1).

Dunn, E. H., and J. R. Sauer. 1997.Monitoring Canadian bird popula-tions with winter counts. Pages 49-55in Dunn, E. H., M. D. Cadman, andJ. Bruce Falls, editors, Monitoringbird populations: the Canadian expe-rience. Canadian Wildlife Service,Occasional Paper 95.

Fewster, R. M., S. T. Buckland, G. M.Siriwardena, S. R. Baillie, and J. D.Wilson. 2000. Analysis of populationtrends for farmland birds using gener-alized additive models. Ecology81:1970–1984.

Link, W. A., and J. R. Sauer 1998.Estimating population change fromcount data: application to the NorthAmerican Breeding Bird Survey.Ecological Applications 8:258-268.

Link, W. A., and J. R. Sauer. 1999a. Onthe importance of controlling foreffort in analysis of count survey data:Modeling population change fromChristmas Bird Count data. Vogelwelt120, Suppl 1:15-20.

Link, W. A., and J. R. Sauer. 1999b.Controlling for varying effort in countsurveys—an analysis of ChristmasBird Count data. Journal ofAgricultural, Biological andEnvironmental Statistics 4:116-125.

Link, W. A., and J. R. Sauer. 2002. Ahierarchical model of populationchange with application to CeruleanWarblers. Ecology (In press).

Nichols, J. D., J. E. Hines, J. R. Sauer,F. W. Fallon, J. E. Fallon, and P. J.Heglund. 2000. A double-observerapproach for estimating detectionprobability and abundance from avianpoint counts. The Auk 117:393-408.

Robbins, C. S., D. Bystrak, and P. H.Geissler. 1986. The Breeding BirdSurvey: its first fifteen years, 1965-1979. U.S. Fish and Wildlife Service,Resource Publication 157.

Root, T. L. 1988. Atlas of WinteringNorth American Birds. University ofChicago Press, Chicago, IL.

Sauer, J. R., S. Schwartz, and B. Hoover.1996. The Christmas Bird CountHome Page. Version 95.1. PatuxentWildlife Research Center, Laurel, MD.http://www.mbr-pwrc.usgs.gov/bbs/cbc.html

Thompson, S. K. 1992. Sampling. JohnWiley & Sons, New York.

14 AMERICAN BIRDS

Page 15: American Birds - National Audubon Society | Audubon

THE 102ND CHRISTMAS BIRD COUNT AMERICAN BIRDS 15

Editor’s note: This article, originallypublished as “Eastcentral Iowa andNorthwestern Illinois Nocturnal andDiurnal Raptor Population Trends: FortyYears of Christmas Bird Count Data,”was reprinted with permission fromHawkwatching in the Americas (Bildsteinand Klem, Eds.) 2001. Pages 181-193.

Abstract

The Quad City Audubon Society hasconducted and compiled five ChristmasBird Counts (CBC) since the mid 1950s.Each of these counts is centered on ornear the Mississippi River in eastcentralIowa and northwest Illinois. Countshave accumulated 40 years of dataregarding avian populations wintering inthe Midwest. Long-term informationconcerning regional raptor populationsis often very minimal. Therefore wedecided to use this CBC data set, gener-ated from a relatively small geographicarea, to examine the observed changes inmidwestern raptor populations since themid 1950s. During our analysis, wepooled the total number of individualsfor each species from all five CBCs eachyear. Furthermore, we combined thedata into five-year increments. Sevenspecies (Sharp-shinned Hawk [Accipiterstriatus], Cooper’s Hawk [A. cooperii],Red-tailed Hawk [Buteo jamaicensis],American Kestrel [Falco sparverius], BaldEagle [Haliaeetus leucocephalus], BarredOwl [Strix varia], and Eastern Screech-Owl [Otus asio]) have been characterizedby populations that were stable untilthe early 1970s to late 1980s, afterwhich they have increased steadily. TheRed-shouldered Hawk (B. lineatus),Long-eared Owl (Asio otus), and Short-earedOwl (Asio flammeus) have undergonesubstantial declines and have not yetrebounded. Northern Harriers (Circus

cyaneus) and Rough-legged Hawks (B.lagopus) fluctuated since the mid 1950s,and Great Horned Owls (Bubo virgini-anus) have steadily increased. Althoughour project included data from a relativelysmall geographic area, our results indicatethat long-term data generated by theCBC program can assist in monitoringregional raptor populations over time.

Introduction

It is extremely difficult to monitorregional populations of raptors. Due totheir ecology, behavior, and natural lowdensities (Newton 1979, Craighead andMindell 1981, Fuller and Mosher 1981,Millsap and LeFranc 1988, Bibby et al.1992, Preston and Beane 1996), mostraptors do not lend themselves well tobeing monitored during the breedingseason by programs such as the BreedingBird Survey, Breeding Bird Atlases, or byMAPS projects (Ralph et al. 1993, Priceet al. 1995, Spess-Jackson et al. 1996).In contrast, during migration, monitor-ing sites along migratory corridors haveproven to be relatively effective in mon-itoring regional raptor population trends(Spofford 1969, Hackman and Henny1971, Nagy 1977, Hussell 1981, Hussell1985, Zalles and Bildstein 2000).Especially effective are sites that concen-trate migrating raptors along “leadinglines” such as mountain ridges andcoastlines (Mueller and Berger 1967,Dunne et al. 1984, Bednarz andKerlinger 1989, Kerlinger 1989, Bibbyet al. 1992, Bildstein et al. 1993, Zallesand Bildstein 2000). Unfortunately, rel-atively few long-term monitoring sitesexist (Bildstein et al. 1998).

The only long-term, large-scale avianmonitoring project that occurs during thewinter season is the Christmas BirdCount (CBC) sponsored by the National

Audubon Society (Bock and Root 1981,Drennan 1981, Arbib 1981, Root 1988).The Quad City Audubon Society hascoordinated and compiled data from fiveCBCs along the Mississippi River in east-central Iowa and northwest Illinois sincethe middle 1950s (McKay 1998, McKay1999). The CBC program dates from1900 when it began as an organized protestby early conservationists to the tradition ofChristmas holiday “hunts” carried out pri-marily by the wealthy (Root 1988). Duringthese events, people shot as much wildlifeas they could in a day, including enor-mous numbers of songbirds.

The CBC program has evolved fromthis early form of protest to become thepremier avian monitoring project dur-ing the early winter period in NorthAmerica. By the late 1900s, between1,700 and 1,800 CBCs were conductedannually across North America. Theprogram has several advantages and dis-advantages closely associated with it.

CBCs possess several sampling biases.Among these are differences in the num-ber of observers, field parties, and hoursof observation effort (Bibby et al. 1992).Count methodologies and strategies alsovary widely. For example, some countshave a large number of small territoriesthat are primarily walked, while othersemploy a small number of larger territo-ries that are driven. Owl tapes are usedon some counts to attract passerines,while other compilers are opposed tothis strategy. Counts also differ on theuse of owl tapes to elicit call-backresponses. Each year between 45,000and 50,000 people participate and con-tribute data to CBCs. This introducesthe problem of extreme differences inobserver abilities to detect and identifythe various avian species (Bibby et al.1992). Additionally, there are no weather

Long-term Trends of Raptors on CBCs in the MidwestKelly J. McKay and Jon W. Stravers

Midwest Raptor Research Fund, 420 1st Avenue, Hampton, IL 61256 USA

Peter C. PetersenQuad City Audubon Society, 1108 Jersey Ridge Road, Davenport, IA 52803 USA

Casey J. Kohrt, Joseph S. Lundh, and Gary V. SwensonU.S. Army Corps of Engineers, Mississippi River Project, Pleasant Valley, IA 52767 USA

Page 16: American Birds - National Audubon Society | Audubon

16 AMERICAN BIRDS

parameters to adhere to. Thus, weathercan significantly influence a count fromyear to year (Bibby et al. 1992). Finally,the CBC program provides a smallamount of overall sampling effort, witheach count only being conducted oneday per year. These biases are inherentwith the CBC program and cause areduction in the scientific validity of thedata (Bibby et al. 1992).

Despite these problems, the CBC pro-gram has several definite advantages.With counts being conducted all acrossNorth America each year, it is one of thefew programs that provides valuableinformation regarding long-term avianpopulation trends over a broad geographicarea (Root 1988). CBCs are semi-stan-dardized, in that the same study area issurveyed each year at the same approxi-mate time and often by the same approx-imate number of observers. The CBCprogram also is extremely cost-effectivesince it uses all volunteer participants(Bibby et al. 1992). In fact, the five dollarper count participation fee allows theCBC program to be financially self-suffi-cient. The program also is highly produc-tive, with thousands of birders on the con-tinent actively participating each year. Thevast geographic area covered, as well as theparticipation and expertise required, pre-cludes a more scientifically and statistical-ly valid project from being conducted(Bibby et al. 1992).

CBCs provide one of the few sourcesof data for monitoring long-term avianpopulation trends. Given the difficultyin monitoring raptor populations, ourobjective was to examine the long-termpopulation trends for 13 raptor speciesin eastcentral Iowa and northwestIllinois, using data from the five CBCscompiled by the Quad City AudubonSociety over a 40-year period. We alsooffered possible explanations for theobserved trends.

Study Area

We analyzed count results from fiveCBCs on or near the Mississippi Riverin eastcentral Iowa and northwestIllinois (Figure 1). All five CBCs are in

the eastcentral Iowa counties of Louisa,Muscatine, Scott, Clinton, and Jackson,and in the northwest Illinois counties ofMercer, Rock Island, Whiteside, andCarroll. Count areas included WesternMercer County (centered at 41˚10’Nand 90˚55’W, Illinois), Muscatine(41˚25’N and 91˚00’W, Iowa),Davenport (41˚31’N and 90˚30’W,Iowa), Princeton-Camanche (41˚45’Nand 90˚21’W, Iowa), and Clinton(41˚58’N and 90˚09’W, Iowa) (NationalAudubon Society 1999).

Methods

CBCs consist of a 15-mile diametercount circle (Bibby et al. 1992). Countcircles usually are divided into severalterritories, each surveyed by an individ-ual field party. Counts are conductedduring a 24-hour period extending frommidnight to midnight. The overall CBCcount period is 16-17 days long, encom-passing the latter half of Decemberthrough the first few days of January(Bibby et al. 1992, National AudubonSociety 1999). During a CBC, the totalnumber of individuals for each speciesidentified is recorded.

Root (1988) and Bibby et al. (1992)recommend examining the data as themean number counted per CBC over10-year periods. They maintain thatusing means and 10-year incrementswill reduce biases due to weather orabnormal bird movements (Bibby et al.1992). However, we believed that thismethod may not adequately detect moresubtle changes in populations, as well asthe historical timing of such changes.Thus, for our analysis we combined thetotal number of individuals recorded foreach species from all five count circles.Additionally, we pooled the data intofive-year increments.

In order for a species to be consideredin our examination, it had to meet twobasic qualifications. First, it had to beconsidered at least a regular winter resi-dent within the study area (Bohlen 1989,Kent and Dinsmore 1996, Petersen1996). Second, the species had to berecorded on at least one of the counts for

a minimum of 25 years. Thirteen diurnaland nocturnal raptors met these criteria.These 13 species were separated into fivegroups based on the general habitat typein which they typically occurred duringthe winter season (Bohlen 1989, Kentand Dinsmore 1996). The first group ofthree species (Sharp-shinned Hawk[Accipiter striatus], Cooper’s Hawk [A.cooperii], and Long-eared Owl [Asio otus])prefers dense mixed deciduous – conifer-ous woodlands, especially pine grovesduring the winter. The second group(American Kestrel [Falco sparverius] andRed-tailed Hawk [Buteo jamaicensis])prefers upland woodland – agriculturaledge habitat, as well as open agriculturalfields. The third group (Great HornedOwl [Bubo virginianus] and EasternScreech-Owl [Otus asio]) is found inupland woodland habitats. The fourthgroup (Rough-legged Hawk [B. lagopus],Northern Harrier [Circus cyaneus], andShort-eared Owl [Asio flammeus]) is typi-cally found in grassland habitats. Theremaining three species (Barred Owl[Strix varia], Red-shouldered Hawk [B.lineatus], and Bald Eagle [Haliaeetus leu-cocephalus] are found in flood plain bot-tomland forest habitat.

Results

The two accipiter species (Sharp-shinned Hawk and Cooper’s Hawk)have exhibited similar population trendsover the past 40 years. Both maintainedlow but stable populations until the lat-ter half of the 1970s, at which time theSharp-shinned population increasedsubstantially (Figure 2). They have

Figure 1. Map illustrating all CBC circlesin Iowa. Circles 3, 4, 5, 7, and 8 representcounts compiled by the Quad CityAudubon Society.

Page 17: American Birds - National Audubon Society | Audubon

THE 102ND CHRISTMAS BIRD COUNT AMERICAN BIRDS 17

continued to increase gradually since.The Cooper’s Hawk populationremained stable until the latter 1980s,when they also demonstrated a substan-tial increase. Cooper’s Hawks also haveincreased since then. Long-eared Owlsoccurred in fairly large numbers throughthe late 1960s, after which the popula-tion substantially declined during theearly portion of the 1970s (Figure 2).This species suffered another severedecline during the early 1980s and hasnot recovered (Figure 2).

The two upland woodland and edgeraptors have exhibited similar popula-tion trends over the 40-year study peri-od (Figure 3). The Red-tailed Hawkpopulation fluctuated somewhat, butremained fairly stable until the latter1980s when it exhibited a rather strongincrease (Figure 3). This increase hascontinued through the early 1990s. TheAmerican Kestrel population was stableuntil the early 1980s, when it began toincrease (Figure 3). Kestrels have contin-ued to exhibit a rather pronouncedincrease in numbers since then.

Great Horned Owls have increasedsteadily throughout the entire 40-yearperiod (Figure 4). The rate of increaseseems to have accelerated slightly sincethe early 1980s. Eastern Screech-Owlswere stable until the early 1970s, whenthey began to increase substantially(Figure 4). They have continued toincrease in the 1990s.

Among the grassland species, Rough-legged Hawk numbers have been char-acterized by relatively large fluctuationsover the last 40 years (Figure 5). TheNorthern Harrier and Short-eared Owlexhibited somewhat similar populationtrends through the 1970s (Figure 5).Both species increased during the late1950s and 1960s, with the Short-earedOwl peaking in the late 1960s and theNorthern Harrier in the early half of the1970s. Likewise, both species experi-enced substantial declines during thelatter 1970s. Since this time, their pop-ulation trends have diverged. NorthernHarrier numbers have stabilized andbegan to increase fairly substantially

Figure 2. Population trends of dense woodland raptors (Sharp-shinned Hawk, Cooper’sHawk, Long-eared Owl).

Figure 3. Population trends of upland woodland and agricultural edge raptors(Red-tailed Hawk, American Kestrel).

Figure 4. Population trends of upland woodland owls (Great Horned Owl, EasternScreech-Owl).

Page 18: American Birds - National Audubon Society | Audubon

18 AMERICAN BIRDS

during the latter 1980s and early 1990s(Figure 5). Short-eared Owl numbershave continued to decline and havenever recovered (Figure 5).

The population trends of Barred Owlsand Red-shouldered Hawks have dif-fered substantially. Barred Owl numberswere stable until the early 1980s whenthey exhibited a strong increase (Figure6a). The increase, which continuedthrough the late 1980s, has since leveledoff. Red-shouldered Hawks “crashed” inthe late 1960s and have yet to recover(Figure 6a). The Bald Eagle populationwas very stable until the latter half of the1970s when it displayed a substantialincrease (Figure 6b). For the most part,the population has continued to increasesubstantially since, particularly duringthe late 1980s and early 1990s.

Discussion

Sharp-shinned and Cooper’s Hawksare secretive accipiters that typicallyoccur in relatively low densities, whichexplains their low numbers on CBCs(Root 1988). The recent recoveries ofthese two accipiters may be linked to thebanning of organochlorine pesticides,like DDT, in 1972 (Bohlen 1989). Manyof their passerine prey species are insec-tivorous. As a consequence, DDT mayhave accumulated to detrimental levels inaccipiters causing reduced reproductivesuccess and associated declines in popu-lations. Much of the recovery of thesetwo species also can be attributed to therapid increase in “bird feeding” by peoplesince the 1970s (DeGraaf and Thomas1974, Root 1988). Bird feeding isundoubtedly aiding the populations ofSharp-shinned and Cooper’s Hawks bymaintaining a larger avian prey base fur-ther north during the winter months(Bohlen 1989, Kent and Dinsmore1996). The delay in the recovery of theCooper’s Hawk population may be relat-ed to the fact that their breeding rangeextends much further south throughoutthe Midwest than does the breedingrange of the Sharp-shinned Hawk(National Geographic Society 1983,Bielefeldt and Rosenfield 1994). The

Cooper’s Hawk population was suspect-ed to be severely impacted by DDT, aswell as extensive loss of nesting habitatprimarily to agricultural development.Long-eared Owl populations begandeclining in the early 1970s and havecontinued this trend ever since. The

overall severe reduction in Long-earednumbers is not well understood (Kentand Dinsmore 1996). However, webelieve this trend is directly linked to thecontinuing loss of dense, moist wood-land habitats throughout much of theMidwest (Bohlen 1989).

Figure 5. Population trends of grassland raptors (Rough-legged Hawk, NorthernHarrier, Short-eared Owl).

(a)

Figure 6. Population trends of flood plain bottomland forest raptors: (a) Barred Owland Red-shouldered Hawk, (b) Bald Eagle.

(b)

Page 19: American Birds - National Audubon Society | Audubon

THE 102ND CHRISTMAS BIRD COUNT AMERICAN BIRDS 19

Red-tailed Hawk and AmericanKestrel populations have increased sub-stantially in recent years (Byre et al.1991, Castrale 1991). The similar trendsexhibited by red-tails and kestrels arethought to be associated with theiradaptability and use of edge-dominatedhabitats (Kent and Dinsmore 1996).Throughout the Midwest, the amount ofupland woodland—agricultural edgehabitat has been increasing steadily. Thisincrease in edge-oriented habitat, whichwas possibly detrimental to Cooper’sHawks and Long-eared Owls, appears tofavor Red-tailed Hawks and AmericanKestrels (Spess-Jackson et al. 1996). TheRed-tailed Hawk has undoubtedly bene-fited from decreased human persecution(Newton 1979). Being a large raptor,often found in open areas, Red-tails werefrequent victims of shootings. Since alarge part of the American Kestrel’s diet insummer consists of insects, this speciesprobably suffered population declinesdue to DDT (Burns et al. 1994).Consequently, kestrels have no doubtbenefited from the banning of this chem-ical. However, we also believe the kestrelpopulation has been assisted in Iowa bythe Department of Natural Resources’nest box program, which has beenextremely successful (Varland et al. 1992).

The steady increase in Great HornedOwls is most likely due to the extremeadaptability of the species. Horned owlsthrive in many habitat types, includingall stages of woodland succession(Dinsmore et al. 1984). Additionally, aswith most of the other species we exam-ined, Great Horned Owls may have ben-efited from decreased human persecu-tion, which has resulted from publiceducation programs and laws designed toprotect predators (Newton 1979). Theincreasing trend in the Eastern Screech-Owl population may be at least partiallylinked to our increasing knowledge ofthe importance of snag habitat. Holes insnags are very important to Screech-Owlsfor roosting and nesting (Bohlen 1989,Spess-Jackson et al. 1996). This speciesprefers smaller, more open, wood lots. Assuch, they have probably benefited from

the widespread fragmentation of largerforests throughout the Midwest (Spess-Jackson et al. 1996). Alternatively, webelieve that a large part of the increase inScreech-Owl numbers is actually theresult of an increasing use of owl audiotapes to elicit call-back responses (Kentand Dinsmore 1996). We began to wide-ly use tapes in the early 1970s. This wasthe same time period that we detected aspike in Eastern Screech-Owl numbers.As a result, the increasing populationtrend in this species may be an artifact ofchanging count methodology more thanan actual increase in the number of birds.

Widely fluctuating numbers of Rough-legged Hawks are not unexpected. This isa tundra-nesting species, which probablyexperiences cyclic migrations as well aspopulation “irruptions” and “crashes”similar to other species of the far north(Bohlen 1989, Kent and Dinsmore1996). These cycles would certainly helpaccount for their fluctuating populationson midwestern CBCs. The severe popu-lation declines noted for NorthernHarriers and Short-eared Owls in the late1970s was attributed to the extensivealteration and loss of grassland habitatsthroughout the Midwest (Hands et al.1989, Martell 1991, Herkert et al. 1993).Interestingly, although Harrier numbersbegan to increase in the late 1980s andearly 1990s, Short-eared Owl numberscontinued to decline. During this sameperiod, there was a substantially largeincrease in the amount of available grass-land habitat due to the ConservationReserve Program (CRP) (Spess-Jacksonet al. 1996). Consequently, NorthernHarriers appear to benefit from CRP land(Spess-Jackson et al. 1996), while Short-eared Owls may not (Martell 1991).

The Barred Owl population was stableuntil it increased substantially during the1980s. A large part of this strong increasemay be due to our growing knowledge ofthe importance of snag habitat. As with theEastern Screech-Owl, holes in snags arecrucial to Barred Owls for nesting androosting sites (Craighead and Craighead1956). However, we also believe that partof the increase was due to an increase in the

use of owl playback tapes by Christmascounters (Kent and Dinsmore 1996).Additionally, as with most other raptors,this species has undoubtedly benefitedfrom decreased human persecution. Incontrast, the Red-shouldered Hawk popu-lation has not recovered from the severe“crash” of the late 1960s. Red-shoulders area species of the flood plain periphery, pre-ferring areas with contiguous flood plainbottomland and upland bluff forest com-ponents (Stravers and McKay 1994).Unfortunately, throughout much ofIllinois and Iowa, the bluff forests haveeither been lost to development or the bluffand flood plain forests have been isolatedfrom each other, primarily by agriculturaland residential development within theflood plain (Stravers 1992, Stravers andMcKay 1994, Spess-Jackson et al. 1996).By comparison, the Bald Eagle populationhas been increasing substantially since thelatter 1970s. We believe the strong recoveryof this species was directly linked to thebanning of DDT, along with the extraemphasis and protection afforded the BaldEagle, which, until recently, was a federallyendangered species (Grier 1988).Throughout the 40-year study period, BaldEagles always maintained impressive num-bers. This was due to the fact that all five ofour CBCs contained large stretches of river,located within one of the major winteringconcentration sites for the species (Millsap1986, McKay 1992).

Our results suggest that CBCs mayprovide valuable information regardinglong-term raptor population trends.Nevertheless, our explanations for popu-lation increases or declines remain specu-lative. More research on the individualspecies is needed. Several species includ-ing the Sharp-shinned Hawk, Cooper’sHawk, American Kestrel, Red-tailedHawk, Barred Owl, Eastern Screech-Owl, and Bald Eagle had relatively stablepopulations which have exhibitedincreases recently. Red-shouldered Hawk,Long-eared Owl, and Short-eared Owlpopulations have experienced severedeclines from which they have yet torecover. Northern Harrier and Rough-legged Hawk populations have fluctuated,

Page 20: American Birds - National Audubon Society | Audubon

20 AMERICAN BIRDS

while Great Horned Owl numbers haveincreased steadily. Finally, we caution thatour results are based on data from onlyfive CBCs. In order to truly examineregional raptor population trends, datafrom a much larger number of CBCsover a broader geographic area must beanalyzed. What we report here is a pre-liminary pilot study.

Acknowledgments

The authors gratefully acknowledgethe helpful recommendations offered byKeith Bildstein, Dan Klem, and JohnSmallwood. Their suggestions assisted inthe preparation of this manuscript. Wealso thank the hundreds of volunteercounters who have participated on ourfive counts over the years. Their efforts,including thousands of hours in thefield, collecting data, have made thesecounts successful. In fact, without theirassistance we would have no data to ana-lyze and interpret. As a result, we dedi-cate this manuscript to all of thoseChristmas counters.

References

Arbib, R.S. 1981. The Christmas BirdCount: constructing an “idealmodel.” Studies in Avian Biology6:30-33.

Bednarz, J.C., and P. Kerlinger. 1989.Monitoring hawk populations bycounting migrants. Pages 328-342 inProceedings of the northeast raptormanagement symposium and work-shop (B.G. Pendleton, Ed.). NationalWildlife Federation Science andTechnical Series, Number 3. NationalWildlife Federation, Washington, D.C.

Bibby, C.J., N.D. Burgess, and D.A.Hill. 1992. Bird census techniques.Academic Press, San Diego, California.

Bielefeldt, J., and R.N. Rosenfield.1994. Summer birds of conifer plan-tations in southeastern Wisconsin.Passenger Pigeon 56:123-135.

Bildstein, K.L., J. Brett, L. Goodrich,and C. Viverette. 1993. Shooting gal-leries. American Birds 47:38-43.

Bildstein, K.L., W. Schelsky, J.I. Zalles,and S. Ellis. 1998. Conservation sta-

tus of tropical raptors. Journal ofRaptor Research 32:3-18.

Bock, C.E., and T.L. Root. 1981. TheChristmas Bird Count and avianecology. Studies in Avian Biology6:17-23.

Bohlen, H.D. 1989. The birds ofIllinois. Indiana University Press,Bloomington, Indiana.

Burns, S.A., W.M. Jarman, T.J. Cade,L.F. Kiff, and B.J. Walton. 1994.Organochlorines and eggshell thin-ning in Peregrine Falcon (Falco pere-grinus) eggs from the eastern UnitedStates, 1986-1988. Pages 709-716 inRaptor conservation today (B.-U.Meyburg and R.D. Chancellor, Eds.).Helm Information Limited, London,United Kingdom.

Byre, V.J., J.M. Siska, and M.F. Spreyer.1991. Falcons. Proceedings of theMidwest raptor management sympo-sium and workshop (B.G. Pendletonand D.L. Krahe, Eds.). NationalWildlife Federation Science andTechnical Series 15:69-78.

Castrale, J.S. 1991. Eastern woodlandbuteos. Proceedings of the Midwestraptor management symposium andworkshop (B.G. Pendleton and D.L.Krahe, Eds.). National WildlifeFederation Science and TechnicalSeries 15:50-59.

Craighead, J.J., and F.C. Craighead.1956. Hawks, owls and wildlife.Stackpole Publishing Company,Harrisburg, Pennsylvania.

Craighead, F.C., Jr., and D.P. Mindell.1981. Nesting raptors in westernWyoming, 1947 and 1975. Journal ofWildlife Management 45:865-872.

DeGraaf, R.M., and J.W. Thomas.1974. A banquet for birds. NaturalHistory 83:40-45.

Dinsmore, J.J., T.H. Kent, D. Koenig,P.C. Petersen, and D.M. Roosa.1984. Iowa birds. Iowa StateUniversity Press, Ames, Iowa.

Drennan, S.R. 1981. The ChristmasBird Count: an overlooked andunderused sample. Studies in AvianBiology 6:24-29.

Dunne, P., D. Keller, and R.

Kochenberger. 1984. Hawk watch: aguide for beginners. Cape May BirdObservatory, and New JerseyAudubon Society, Cape May, NewJersey.

Fuller, M.R., and J.A. Mosher. 1981.Methods of detecting and countingraptors. Studies in Avian Biology6:235-246.

Grier, J.W. 1988. Northern states BaldEagle recovery team. Technical report,Raptor Research Foundation,Minneapolis, Minnesota.

Hackman, C.D., and C.J. Henny.1971. Hawk migration over WhiteMarsh, Maryland.Chesapeake Science 12:137-141.

Hands, H.M., R.D. Drobney, andM.R. Ryan. 1989. Status of theNorthern Harrier in the northcentralUnited States. Technical report, U.S.Fish and Wildlife Service, TwinCities, Minnesota.

Herkert, J.R., R.E. Szafoni, V.M. Kleen,and J.E. Schwegman. 1993. Habitatestablishment, enhancement andmanagement for forest and grasslandbirds in Illinois. Illinois Departmentof Conservation, Natural HeritageTechnical Publication Number 1.

Hussell, D.J.T. 1981. The use of migra-tion counts for monitoring bird pop-ulation levels. Studies in AvianBiology 6:92-102.

Hussell, D.J.T. 1985. Analysis of hawkmigration counts for monitoringpopulation levels. Pages 243-254 inProceedings of hawk migration con-ference IV (M. Harwood, Ed.).Hawk Migration Association ofNorth America, Lynchburg, Virginia.

Kent, T.H., and J.J. Dinsmore. 1996.Birds in Iowa. Kent and DinsmorePublishers, Iowa City, Iowa.

Kerlinger, P. 1989. Flight strategies ofmigrating hawks. University ofChicago Press, Chicago, Illinois.

Martell, M.S. 1991. Grassland owls.Proceedings of the Midwest raptormanagement symposium and work-shop (B.G. Pendleton and D.L. Krahe,Eds.). National Wildlife FederationScience and Technical Series 15:96-104.

Page 21: American Birds - National Audubon Society | Audubon

THE 102ND CHRISTMAS BIRD COUNT AMERICAN BIRDS 21

McKay, K.J. 1992. Bald Eagle popula-tion dynamics and the influence offoraging perch habitat quality on dis-tribution in west-central Illinois. M.S.thesis, Western Illinois University,Macomb, Illinois.

McKay, K.J. 1998. Summary of the1997 Christmas Bird Counts.Mississippi Currents, Newsletter of theQuad City Audubon Society 9(2):4.

McKay, K.J. 1999. Highlights of the1998 Christmas Bird Counts.Mississippi Currents, Newsletter of theQuad City Audubon Society 10(3):3.

Millsap, B.A. 1986. Status of winteringBald Eagles in the conterminous 48states. Wildlife Society Bulletin14:433-440.

Millsap, B.A., and M.N. LeFranc, Jr.1988. Road transects for raptors: howreliable are they? Journal of RaptorResearch 22:8-16.

Mueller, H.C., and D.D. Berger. 1967.Wind drift, leading lines, and diurnalmigration. Wilson Bulletin 79:50-63.

Nagy, A.C. 1977. Population trendindices based on 40 years of autumncounts at Hawk Mountain Sanctuaryin northeastern Pennsylvania. Worldconference on birds of prey (R.D.Chancellor, Ed.). InternationalCouncil for Bird Preservation.

National Audubon Society. 1999.American Birds (Ninety-ninthChristmas Bird Count [1998-1999]).National Audubon Society, New York,New York.

National Geographic Society. 1983. Fieldguide to the birds of North America.National Geographic Society,Washington, D.C.

Newton, I. 1979. Population ecology ofraptors. Buteo Books, Vermillion,South Dakota.

Petersen, P.C. 1996. A field list of birdsof the Quad City region. Quad CityAudubon Society, Davenport, Iowa.

Preston, C.R., and R.D. Beane. 1996.Occurrence and distribution of diur-nal raptors in relation to humanactivity and other factors at RockyMountain Arsenal, Colorado.Pages 365-374 in Raptors in human

landscapes (D. Bird, D. Varland, andJ. Negro, Eds.). Academic Press, SanDiego, California.

Price, J., S. Droege, and A. Price. 1995.The summer atlas of North Americanbirds. Academic Press, San Diego,California.

Ralph, C.J., G.R. Geupel, P. Pyle, T.E.Martin, and D.F. DeSante. 1993.Handbook of field methods for mon-itoring landbirds. United StatesDepartment of Agriculture, Generaltechnical report, PSW-GTR-144,Albany, California.

Root, T.L. 1988. Atlas of winteringNorth American birds. An analysis ofChristmas Bird Count data.University of Chicago Press, Chicago,Illinois.

Spess-Jackson, L., C.A. Thompson, J.J.Dinsmore, B.L. Ehresman, J.Fleckenstein, R. Cecil, L.M.Hemesath, and S.J. Dinsmore. 1996.The Iowa breeding bird atlas.University of Iowa Press, Iowa City,Iowa.

Spofford, W.R. 1969. Hawk Mountaincounts as population indices innortheastern America. Pages 323-332in Peregrine Falcon populations: theirbiology and decline (J.J. Hickey, Ed.).University of Wisconsin Press,Madison, Wisconsin.

Stravers, J.W. 1992. Surveys for nestingRed-shouldered Hawks within pools9-11 and 16-19 of the UpperMississippi River Valley. Technicalreport, Iowa Department of NaturalResources, Des Moines, Iowa.

Stravers, J.W., and K.J. McKay. 1994.Status of the Red-shouldered Hawkwithin the Upper Mississippi RiverValley and management guidelinesfor nesting habitat. Technical report,U.S. Fish and Wildlife Service,Winona, Minnesota, Quincy, Illinois,and Minneapolis, Minnesota.

Varland, D., R.D. Andrews, and B.L.Ehresman. 1992. Establishing a nestbox program for American Kestrelsalong an interstate highway. Technicalreport, Iowa Department ofTransportation, Ames, Iowa, and

Iowa Department of NaturalResources, Boone, Iowa.

Zalles, J.I., and K.L. Bildstein (Eds.).2000. Raptor watch: a global directo-ry of raptor migration sites. BirdLifeConservation Series Number 9.BirdLife International, Cambridge,United Kingdom, and HawkMountain Sanctuary, Kempton,Pennsylvania.

Page 22: American Birds - National Audubon Society | Audubon

22 AMERICAN BIRDS

Editor’s note: This article, originallypublished as “2001 Yellowstone ChristmasBird Count Results,” was reprinted withpermission from Yellowstone Science (Vol.10, No. 1), Winter 2002 to show theworkings of an individual count circle. Amore detailed summary of past YellowstoneChristmas Bird Count results and methodscan be found in the Winter 2001 issue ofYellowstone Science (Vol. 9, No. 1). Wehope to explore changes over time in theYellowstone National Park count circle in afuture issue of American Birds.

On December 16, 2001, theYellowstone Christmas Bird Count wasconducted in the Gardiner, Montana—Mammoth, Wyoming, area. This YCBCmarks the 29th year for this traditionalbird survey. The established center pointfor this bird count is the North Entranceof Yellowstone National Park andextends 7.5 miles from this point in anydirection, with boundary limits basicallyeast to Blacktail Ponds, north to themining town of Jardine, Montana, andnorthwest to Corwin Springs, Montana.The YCBC is divided into teams ofobservers to maximize landscape cover-age. All bird species and total individualbirds detected during the count day areincluded in the final results. Additionalbirds incidentally observed three daysbefore and three days after official countday are included in another categorycalled the count week totals.

Since the YCBC is totally voluntary,the number of observers showing up inany given year is never known untilcount day. However, each year at least ahalf dozen skilled observers repeatedlyreturn to participate in the YCBC.Weather conditions highly influenceoverall participant turnout as do person-al holiday plans. The number of peopleparticipating in the YCBC has littlebearing on the number of bird species orindividuals detected during count day.In fact, weather plays a greater role infinding birds than does the number of

participants. Because of access limita-tions in the winter, experience hasshown birds can be best counted in spe-cific habitats. The more inclement thewinter weather (e.g., cold temperaturesand deep snows) the better the birding,since birds are concentrated primarilynear bird-feeding stations, riparian areas,

and geothermal or open water areas.Birds are also less concentrated duringmild weather conditions, since naturalfoods are more available. Ironically, thelargest number of participants show upduring years of mild weather conditionswhen birding is just average or belowaverage (Figure 1). Hence, mild weather

Yellowstone Christmas Bird CountDecember 16, 2001

Species Yell.-Wy. Yell.-Mt. Outside TotalsYell. N.P.- Mt.

Green-winged Teal 25 11 36

Mallard 51 66 117

Barrow’s Goldeneye 12 12

Common Merganser 5 5

Bald Eagle 6 5 6 17

Rough-legged Hawk 1 1

Golden Eagle 2 2 4

Common Snipe 2 2

Rock Dove 26 22 48

Belted Kingfisher 1 1 2

Northern Flicker 1 3 4

Horned Lark 1 1

Gray Jay 4 4

Steller’s Jay 1 4 5

Pinyon Jay 35 35

Clark’s Nutcracker 39 22 61

Black-billed Magpie 63 8 46 117

Common Raven 49 13 58 120

Black-capped Chickadee 3 8 11

Mountain Chickadee 50 21 71

Red-breasted Nuthatch 9 10 19

Marsh Wren 2 2

American Dipper 10 28 4 42

Townsend’s Solitaire 22 5 20 47

Bohemian Waxwing 19 50 69

Yellow-rumped Warbler 2 2

Song Sparrow 4 3 7

Gray-crowned Rosy Finch 120 120

Black Rosy Finch 2 2

House Finch 36 36

Common Redpoll 75 314 389

Red Crossbill 17 17

Pine Siskin 55 60 115

House Sparrow 15 120 135

Totals 552 152 971 1675

Total Species: 34 (Additional Species Count Week: 2)

Yellowstone National Park: A Look at an Individual Count CircleTerry McEneaney

P.O. Box 168, Yellowstone National Park, WY 82190

Page 23: American Birds - National Audubon Society | Audubon

THE 102ND CHRISTMAS BIRD COUNT AMERICAN BIRDS 23

years for the YCBC result in few birdrarities being detected.

The 2001 Yellowstone ChristmasBird Count tallied a total of 34 birdspecies and 1675 individual birds. Themild weather conditions resulted in anaverage number of species and a slightlyabove average number of individualbirds observed. As expected due to themild weather, a record of 22 observersshowed up for the 2001 YCBC tyingthe previous record set in 1999, anoth-er mild year. Temperatures during the2001 YCBC ranged from 12-27 degreesF., with 3-12 inches of snow, dependingon the elevation, and the edge of therivers were not even frozen.

Three bird records were broken duringthe 2001 YCBC. A total of 389 CommonRedpoll were detected in the count areathis year, compared to the previous recordof 148 set in 1989. The irruption of

Common Redpolls was the result of arare superabundance of food, namelyDouglas-fir seed cones, alder catkins, andexposed grass seed heads, coupled with anearly winter storm forcing redpolls intothe area in November. Two Yellow-rumped Warblers were also detected,whereas in the past only one was seen in1983, 1987, and 1990. Additionally,seven Song Sparrows were found this yearcompared to the previous record of sixobserved in 1988. Two Marsh Wrens werealso found during the 2001 YCBC; thisties the record set last year. Four NorthernFlickers were found in 2001, tying theprevious record set in 1987. Interestinglyenough, species that are regularly detectedsuch as the Common Goldeneye, HairyWoodpecker, Downy Woodpecker, Dark-eyed Junco and American Tree Sparrowcould not be located due to the mild win-ter weather conditions.

In conclusion, a grand total of 95 specieshave been recorded on the YCBC (97species with the YCBC and count weekcombined) during the 29 years the counthas taken place. This year, mild weatherconditions resulted in an average numberof bird species detected, and a slightlyabove average number of individualsobserved. Tolerable weather conditions alsoresulted in tying a record number of 22participants attending the count. However,experience has shown that colder tempera-tures and above average snow depths arethe optimum times for finding the greatestbird richness and abundance during theYCBC. Participants are reminded of thesefactors when deciding on attending futureYCBC’s. Regardless, the YellowstoneChristmas Bird Count tradition continuesand a fun time was had by all.

Figure 1. Number of participants

Page 24: American Birds - National Audubon Society | Audubon

24 AMERICAN BIRDS

An “exotic” species is one that occurs ina region to which it is not native. Exceptfor those few that are (or may eventuallybe) “countable” on personal lists, exoticsare ignored by most birders. Likewise,few ornithologists have shown an interestin exotic species, despite several fertiletopics of study. Lately however, interest isrising, as the distribution, numbers, anddiversity of exotics continue to increase,and impacts to native species (if any)need to be assessed. Here, I demonstratehow CBC data can be used to analyzepopulation trends of exotic birds. Thispaper barely skims the surface of theinformation available. Exotics are mostwidespread in southern California, south-eastern Florida, and throughout the mainHawaiian islands, states that share mildclimates, urban areas extensively land-scaped with exotic vegetation, and majorports of entry for imported wildlife. Mostof the species present in California andFlorida are the result of accidental escapesof cagebirds, but dozens of species havebeen deliberately released into Hawaii—Long (1981) documented 75 species ofgamebirds alone! This paper is biasedtoward California and Florida becausedata from several studies were easily available.Population trends of gamebirds can be dif-ficult to track because of the possibility ofcontinual releases by state agencies and pri-vate hunting clubs. Numbers of doves,parrots, and other cagebirds may also besupplemented by accidental or intentionalreleases, but these likely are much smallerin scale, especially in recent years.

Populations of exotics often experi-ence rapid increases in numbers followedby severe declines. A few species, such asthe Black Francolin (Francolinus francol-inus) in Florida and Louisiana, and theBlue-gray Tanager (Thraupis episcopus)and Java Sparrow (Padda oryzivora) inFlorida, have died out completely.Because of such population fluctuations,large-scale, long-term monitoring ofexotics is needed. CBC data provide sev-

eral advantages for monitoring thesepopulations: 1) counts have been con-ducted in some areas for decades, allow-ing long-term trends to be determined;2) counts occur throughout the UnitedStates and Canada, allowing for large-scale monitoring; 3) CBC circles are allthe same size, and survey methodology isstandardized; 4) CBCs quantify observereffort, thereby allowing direct comparisonof counts that differ widely in the numberof participants or daylight hours; 5) mostCBC circles are located in urban or sub-urban areas, to which most exotics arerestricted; 6) exotics generally are seden-tary, so CBC data should be as valuableas breeding-season surveys; and 7) CBCparticipants are expected to report allbirds seen, even those that are not“countable” on personal lists.

Despite these advantages, severalcaveats need to be mentioned when usingthe CBC database to track exotics: 1) con-fusing genera, such as Streptopelia doves,Amazona parrots, Aratinga parakeets,Euplectes bishops, and Lonchura munias,are poorly known to most participantsand inadequately covered by most fieldguides, creating the potential for errors ofidentification. Recently for example, someFlorida CBCs have listed dozens ofRinged Turtle-Doves (Streptopelia ‘riso-ria’ ) where none were known to occur pre-

viously (i.e., outside of St. Petersburg), andthese almost certainly representmisidentified Eurasian Collared-Doves;2) the reporting of exotics on CBCs isinconsistent and depends on the willing-ness of participants to count them and ofcompilers and editors to include them.Two examples involve the Rock Dove(Columbia livia), which has been presentin North America for nearly 400 years, butwas not widely reported on CBCs untilDecember 1974, and the Muscovy Duck(Cairina moschata), which is locally abun-dant in Florida but was not listed onCBCs in the state until this season; 3) pop-ulations of some exotics, such as thePurple Swamphen (Porphyrio porphyrio)and Red-whiskered Bulbul (Pycnonotusjocosus) in Florida, are entirely outsideCBC circles, while the ranges of probablyall other species are only partially coveredby CBCs; and 4) the online database con-tains errors (e.g., Pranty 2001 and below),so researchers should verify all data fromthe CBC issues of Audubon Field Notes,American Birds, and Field Notes.

Graphs in this paper were created from thedatabase accessible from the Audubonwebsite (www.audubon.org/bird/cbc).These graphs are based on the number ofbirds observed per party-hour, which allowsfor comparison among CBC circles andyears with variable observer effort. Scientific

The Use of Christmas Bird Count Data to Monitor Populations of Exotic BirdsBill Pranty

Audubon of Florida, 410 Ware Boulevard, Suite 702, Tampa, Florida 33619; [email protected]

Figure 1.

Page 25: American Birds - National Audubon Society | Audubon

THE 102ND CHRISTMAS BIRD COUNT AMERICAN BIRDS 25

studies must analyze the data statistically toensure the validity of the results. Because thegraphs presented here have not been ana-lyzed, most of the species selected for thispaper are under formal study.

Eurasian Collared-Dove(Streptopelia decaocto)

The colonization of North America bythis species since the early to mid-1980swell illustrates how CBC data can trackexpansion of a highly successful exotic(Figure 1). Birds colonized southeasternFlorida in the late 1970s or early 1980sfrom a population accidentally releasedinto the Bahamas (Smith 1987). Collared-doves now occur west of the MississippiRiver and continue to increase in range andnumbers (LeBaron 1999, Romagosa andMcEneaney 1999, Romagosa and Labisky2000). Concerns about expanding col-lared-dove populations on native doveswere mentioned by Steadman (1998),Romagosa and McEneaney (1999), andRomagosa and Labisky (2000), but trueimpacts remain to be quantified.

Spotted Dove(Streptopelia chinensis)

This native of southeast Asia wasreleased into southern California around1915 as a gamebird. Although it is local-ly abundant and range expansion contin-ues, numbers at Los Angeles andPasadena began to decline in the late1970s. Causes of the population declinemay be related to increased urbanizationand loss of agricultural and native habi-tats (Johnston and Garrett 1994, Garrettand Walker 2001). Overall numberswithin CBC circles show a steady declinesince the early 1980s (Figure 2).

Budgerigar(Melopsittacus undulatus)

Pranty (2001) used CBC data tograph the startling “boom and bust” pat-tern (Figure 3) of this popular Australiancagebird. Numbers on CBCs in west-central Florida increased from 269 para-keets during 1972–1973 to 2910 indi-viduals two years later. The CBC popu-lation peaked at 6895 individuals during1977–1978 but had crashed to 385parakeets ten years later. Since December1995, Budgerigars have been limited to

Figure 2.

Figure 3.

Figure 4.

Page 26: American Birds - National Audubon Society | Audubon

26 AMERICAN BIRDS

two CBC circles, and numbers haveranged between 43 and 92 individuals.The primary cause of the decline wasthought to be nesting competition withHouse Sparrows (Pranty 2001).

Monk Parakeet(Myiopsitta monachus)

This native of South America is themost successful parrot in North America,aided by its nest-building behavior (whichallows it to avoid competition with nativeand exotic cavity-nesting species), and itsability to survive sub-freezing tempera-tures. Van Bael and Pruett-Jones (1996)used CBC data to document that the U.S.population was doubling every 4.8 years,an exponential increase. Recent CBC dataconfirm that the population continues toincrease (Figure 4). Eradication programsduring 1972–1975 killed dozens of para-keets in several states (Neidermyer andHickey 1977). These programs were suc-cessful at reducing small populations, butprobably would be ineffective today.Eradication efforts in one Argentineanprovince between 1958 and 1960 killed427,206 Monk Parakeets with no appar-ent reduction in the overall population(Neidermyer and Hickey 1977). Floridacontains the bulk of the U.S. population,with 3015 individuals tallied on CBCsthis past season, followed by 908 parakeetsin Connecticut, and dozens of individualsin other states. Further increases in rangeand numbers are likely.

Black-hooded Parakeet(Nandayus nenday)

This currently “uncountable” SouthAmerican native bears watching in Florida,where it is the second most widespread andabundant parrot. Pranty and Lovell (inprep.) used CBC data and other observa-tions to document a significant, long-termpopulation increase (Figure 5), especially inthe St. Petersburg area. Smaller popula-tions are found in southern California.

European Starling(Sturnus vulgaris)

This species probably is the most suc-cessful, widespread, and abundant exoticbird in North America. Starlings colonizedthe continent from 100 birds released atNew York City in 1890–1891. Within a

Figure 5.

Figure 6.

Figure 7.

Page 27: American Birds - National Audubon Society | Audubon

THE 102ND CHRISTMAS BIRD COUNT AMERICAN BIRDS 27

century, their population was estimated at200 million individuals, and breeding pop-ulations were distributed essentially contin-uously from Alaska and Newfoundlandsouth to Baja California and Florida (Cabe1993). CBC data show great variation inannual numbers (probably dependentupon whether large winter roosts are insideor outside CBC circles), but numbers ofbirds observed per party-hour (Figure 6)suggest a population decrease exceeding 66percent in the past 30 years.

Crested Myna(Acridotheres cristatellus)

One or two pairs of mynas from HongKong were released into the Vancouver,British Columbia, area between 1894 and1897. By the 1930s, the population hadsurpassed 5000 individuals, but haddeclined over 50 percent by 1959 (Johnsonand Campbell 1995). CBC data (Figure 7)show a dramatic decline beginning in theearly 1970s, with numbers on CBCs drop-ping below 100 individuals by the mid-1990s. Since December 1995, only one tofour mynas have been found annually—allon the Vancouver CBC—and the species isnearly extirpated. Johnson and Campbell(1995) suggested several causes of the pop-ulation decline, including increased nestingcompetition with European Starlings, lossof foraging habitat from urbanization, andmaladaptation to the local climate.

Spot-breasted Oriole(Icterus pectoralis)

This native of Central America escapedfrom a Miami, Florida, tourist facility in1948 and began to increase in range andnumbers. However, in the 1970s the pop-ulation declined rapidly and has not recov-ered. Robertson and Woolfenden (1992)attributed the decline to a series of severefreezes, which began in January 1977, thatkilled much of the exotic vegetation onwhich the orioles fed. However, CBC data(Figure 8) indicate that the populationdecline began two years earlier. Since thelate 1970s, numbers on Florida CBCs havetotaled less than 60 individuals annually.

House Finch(Carpodacus mexicanus)

This native of western North Americawas accidentally released into New York

Figure 8.

Figure 9.

Figure 10.

Page 28: American Birds - National Audubon Society | Audubon

28 AMERICAN BIRDS

City in 1940 and has been expanding itsrange since that time. The populationincrease, one of the most successful col-onizations in the New World, has beendocumented by several authors (e.g.,Bosakowski 1986 and references therein).During 1994–1996, an epidemic killedtens of millions of House Finches in theeastern United States (Nolan et al. 1998),and CBC data from New York (Figure 9)clearly show its impact on the population.

House Sparrow(Passer domesticus)

Another one of the most successfulexotics in North America, this Old Worldspecies colonized the continent from 100individuals released at New York City in1851–1852 (Lowther and Clink 1992).By the early 1940s, the population wasestimated at 150 million birds. CBC data(Figure 10) show a very gradual declineover the past 40 years, with numbers dur-ing the 1990s less than half of those dur-ing the 1960s. (An error in the databasewas corrected to produce this graph accu-rately). Kricher (1983) noted that HouseSparrow populations in the northeasternUnited States decreased as House Finchpopulations increased. However, HouseSparrow numbers also declined (thoughless severely) in the southeastern UnitedStates, where House Finches had not yetcolonized. Lowther and Clink (1992)attributed the sparrow’s decrease primarilyto the decline of horse transportation, and,since the 1960s, to an increase in largefarms and monoculture crop production.

Christmas Bird Count data can greatlyaid the monitoring of local, regional, andnational populations of exotic birds.However, consistent reporting of exotics is amust, and their identification needs to begiven the same degree of study and docu-mentation as native rarities (Garrett 1993).Participants should not try to “force” anidentification based upon some perceivedlikelihood of occurrence, but rather shouldbe willing to invest in reference materials andtime to hone their identification skills. Insome cases, this means going beyond stan-dard North American guides, as well asbringing along a camera to document obser-

vations. Finally, CBC compilers shouldencourage participants to pay closer atten-tion to all exotics found in their area, not justthose that are officially “countable.”

Acknowledgments

I thank Andrew Farnsworth for assis-tance with the graphs, Kimball Garrettfor helpful comments and for providingselected references, Geoff LeBaron forsuggesting that I write this paper, andHolly Lovell for support.

Literature CitedBosakowski, T. 1986. Winter popula-

tion trends of the House Finch andecologically similar species in north-eastern New Jersey. American Birds40: 1104–1110.

Cabe, P.R. 1993. European Starling(Sturnus vulgaris). In The Birds ofNorth America, No. 48 (A. Poole andF. Gill, eds.). The Academy ofNatural Sciences, Philadelphia, PA;and The American Ornithologists’Union, Washington, D.C.

Garrett, K.L. 1993. Canary-wingedParakeets: the southern Californiaperspective. Birding 25: 430–431.

Garrett, K.L., and R.L. Walker. 2001.Spotted Dove (Streptopelia chinensis).In The Birds of North America, No.586 (A. Poole and F. Gill, eds.). TheBirds of North America, Inc.,Philadelphia, PA.

Johnson, S.R., and R.W. Campbell.1995. Crested Myna (Acridotherescristatellus). In The Birds of NorthAmerica, No. 157 (A. Poole and F.Gill, eds.). The Academy of NaturalSciences, Philadelphia, PA; and TheAmerican Ornithologists’ Union,Washington, D.C.

Johnston, R.F., and K.L. Garrett. 1994.Population trends of introduced birdsin western North America. Pages221–231 in A Century of Change inWestern North America (J.R. Jehl, Jr.,and N.K. Johnson, eds.). Studies inAvian Biology, No. 15. CooperOrnithological Society.

Kricher, J. 1983. Correlation betweenHouse Finch increase and HouseSparrow decline. American Birds 37:358–360.

LeBaron, G.S. 1999. Invasions, irrup-tions, and trends: the Christmas Bird

Count database. North AmericanBirds 53: 217–219.

Long, J.L. 1981. Introduced Birds of theWorld: The Worldwide History,Distribution, and Influence of BirdsIntroduced to New Environments. A.H. &A.W. Reed Pty. Ltd., Sydney, Australia.

Lowther, P.E., and C.L. Clink. 1992.House Sparrow (Passer domesticus). InThe Birds of North America, No. 12(A. Poole, P. Stettenheim, and F. Gill,eds.). The Academy of NaturalSciences, Philadelphia, PA; and TheAmerican Ornithologists’ Union,Washington, D.C.

Neidermyer, W.J., and J.J. Hickey.1977. The Monk Parakeet in theUnited States, 1970–75. AmericanBirds 31: 273–278.

Nolan, P.M., G.E. Hill, and A.M.Stoehr. 1998. Sex, size, and plumageredness predict House Finch survivalin an epidemic. Proceedings of theRoyal Society of London, Series B265: 961–965.

Pranty, B. 2001. The Budgerigar inFlorida: rise and fall of an exotic psittacid.North American Birds 55: 389–397.

Pranty, B., and H.W. Lovell. In prep.Status, distribution, and possible estab-lishment of the Black-hooded Parakeet(Nandayus nenday) in Florida.

Robertson, W.B., Jr., and G.E.Woolfenden. 1992. Florida BirdSpecies: An Annotated List. FloridaOrnithological Society SpecialPublication No. 6, Gainesville, FL.

Romagosa, C.M., and R.F. Labisky.2000. Establishment and dispersal ofthe Eurasian Collared-Dove inFlorida. Journal of Field Ornithology71: 159–166.

Romagosa, C.M., and T. McEneaney.1999. Eurasian Collared-Dove in NorthAmerica and the Caribbean. NorthAmerican Birds 53: 348–353.

Smith, P.W. 1987. The EurasianCollared-Dove arrives in the Americas.American Birds 41: 1370–1379.

Stedman, S.J. 1998. Changing seasons:the nesting season, June 1–July 31,1998. Field Notes 52: 424–426.

Van Bael, S., and S. Pruett-Jones. 1996.Exponential population growth ofMonk Parakeets in the United States.Wilson Bulletin 108: 584–588.

Page 29: American Birds - National Audubon Society | Audubon

THE 102ND CHRISTMAS BIRD COUNT AMERICAN BIRDS 29

The Audubon WatchList 2002 identi-fies North American bird species that needour help. WatchList species are those withpopulation declines, limited geographicrange, and/or threats such as habitat loss onbreeding and wintering grounds. A center-piece of conservation at Audubon, theWatchList focuses attention on at-risk birdspecies so that limited resources are spentwhere they are most needed.

Audubon’s WatchList 2002, an updateof the WatchList presented on the

Audubon website in 1998, for the firsttime includes birds of Hawaii and PuertoRico. Because the list is an independent,science-based, and unbiased evaluation, itincludes an assessment of all bird species,regardless of whether or not they are list-ed as endangered or threatened under theFederal Endangered Species Act.

The WatchList message is clear: Wemust harness the energy of individualsand groups to work collaboratively forbirds and their habitats. By doing so, we

can save birds, their habitats, and theother wildlife that depend on thosehabitats, for the enjoyment and benefitof ourselves and of our children.

Until recently, WatchList criteriahave been primarily based upondeclines as shown by breeding-seasondata. Population trends in other seasonsare important for assessment of the sta-tus of species as well, and the ChristmasBird Count provides the primary tool tolook at early-winter bird populationtrends on a continentwide basis. Datacollected by CBC participants via thestandardized methodology of the count,always in the same areas and at the sametime of year (and often by the samepeople), may show long-term trends inthe species censused.

While declines in wetland and grasslandspecies in many areas have been muchstudied and well documented, shrublandspecies have received considerably lessattention. Many shrubland species are alsoshowing consistent declines as transitionalthicket habitats are either lost to naturalsuccession or developed for use by humans.Here I present preliminary analyses ofChristmas Bird Count data for severalshrubland species included on the currentAudubon WatchList 2002—Long-billedThrasher (Toxostoma longirostre), Bendire’sThrasher (Toxostoma bendirei), CaliforniaThrasher (Toxostoma redivivum), and LeConte’s Thrasher (Toxostoma lecontei).

Long-billed Thrasher occurs indense thickets, brush, and tangles fromsouthern Texas south into easternMexico. Climatic variability (e.g., tem-peratures, precipitation) probablyinfluences abundance and distributionof this species at the northern limits ofits range in Texas. The accompanyinggraph generated using the on-line CBCdata output tool shows data in birds perparty hours from Texas CBCs between1960 and 2000. Note the possible man-ifestation of climatic variability in thelarge fluctuations in birds per party

WatchList Species as Viewed Through the Christmas Bird Count DatabaseAndrew Farnsworth

National Audubon Society, 545 Almshouse Road, Ivyland, PA 18974; [email protected]

Long-billed Thrasher (Toxostoma longirostre). Photo/A. Papadatos/VIREO

Page 30: American Birds - National Audubon Society | Audubon

30 AMERICAN BIRDS

hours among years. The trend for thesedata suggests a slight increase duringthis period, although this trend is notsignificant (p-value = 0.3326, n = 41).The peak in 1975 represents larger-than-average numbers of thrashersreported on several count circles—seven count circles reported at least 30thrashers that year.

Bendire’s Thrasher occurs locally indesert grasslands of the southwesternUnited States. This species has apparent-ly experienced declines in some parts ofits range, possibly a result of overharvest-ing of yuccas, overgrazing of desertgrassland habitats, increasing agriculturalactivity, urban sprawl, or exclusion byCurve-billed Thrasher (Toxostoma curvi-rostre). Using the “Historical Results”link on the CBC home page, quickanalysis of CBC data is an easy way toidentify areas of decline within a species’range. The accompanying graph showsdata in birds per party hours fromArizona CBCs between 1960 and 2000.The trend for these data shows a declineduring this time period that is significant(p-value < 0.0001, n = 41). However,this species is apparently migratory inmuch of its U.S. range, and moreresearch is necessary to evaluate theextent of its decline.

California Thrasher is a highlylocalized species that occurs in chapar-ral, dense thickets, and brushy habitatin central and southern California andadjacent Baja California.

Bendire’s Thrasher (Toxostoma bendirei). Photo/R.& N. Bowers/VIREO

California Thrasher (Toxostomaredivivum). Photo/H.P. Smith Jr./VIREO

Page 31: American Birds - National Audubon Society | Audubon

THE 102ND CHRISTMAS BIRD COUNT AMERICAN BIRDS 31

Because species with limited rangesare perhaps at greater risk for popula-tion declines, quick and easy methodsfor analyzing population trends couldbe very useful for highlighting areas orspecies of concern. This preliminaryanalysis of data in birds per party hoursfrom California CBCs between 1960and 2000 shows how CBCs can be usedin this way. These data on CaliforniaThrasher suggest that this species hasnot exhibited significant declines since1960 (p-value = 0.5446, n = 41).

Le Conte’s Thrasher occurs in aridand sparsely vegetated desert habitats ofsouthern Nevada, southwestern Arizona,southeastern California, and in the adja-cent Mexican states of Baja Californiaand Sonora. The WatchList highlights aneed for more research on this species,because Breeding Bird Survey data maynot accurately reflect its true populationtrends (e.g., thrasher breeding season rel-ative to timing of data acquisition).Again, CBC data can be used for quickand easy analysis of trends over time.The trend for this species fromCalifornia CBCs between 1960 and2000 shows a slight decline since 1960,although this trend is not significant (p-value = 0.1817, n = 41). The large spikein the graph that occurs in the late 1960srepresents the effects of a single countcircle that reported large numbers ofthrashers during its brief existence (Taft-Maricopa CBC, 1968-1971). Le Conte’s Thrasher (Toxostoma lecontei). Photo/J. Fuhrman/VIREO

Page 32: American Birds - National Audubon Society | Audubon

32 AMERICAN BIRDS

The 102nd Christmas Bird CountAlphabetical Index to Regional Summaries

CANADAAlberta 38

British Columbia 41

Manitoba 38

New Brunswick 33

Newfoundland 33

Northwest Territories 38

Nova Scotia 33

Nunavut 38

Ontario 35

Prince Edward Island 33

Quebec 34

Saskatchewan 38

St. Pierre et Miquelon 33

Yukon Territory 41

CARIBBEANBahamas 93

Bermuda 93

Dominican Republic 93

Puerto Rico 93

Virgin Islands 93

LATIN AMERICABelize 90

Brazil 91

Colombia 91

Costa Rica 91

Drake Passage 91

Ecuador 91

Mexico 90

Panama 91

Paraguay 91

Peru 91

Trinidad 91

PACIFIC ISLANDSCommonwealth of Northern Mariana Islands 88

Guam 88

Pacific Islands 88

UNITED STATESAlabama 62

Alaska 43

Arizona 82

Arkansas 73

California 85

Colorado 80

Connecticut 44

Delaware 51

Florida 57

Georgia 56

Hawaii 88

Idaho 79

Illinois 68

Indiana 65

Iowa 72

Kansas 75

Kentucky 60

Louisiana 74

Maine 44

Maryland 51

Massachusetts 44

Michigan 64

Minnesota 70

Mississippi 62

Missouri 73

Montana 79

Nebraska 75

Nevada 82

New Hampshire 44

New Jersey 49

New Mexico 82

New York 47

North Carolina 54

North Dakota 75

Ohio 60

Oklahoma 75

Oregon 84

Pennsylvania 49

Rhode Island 44

South Carolina 55

South Dakota 75

Tennessee 62

Texas 77

Utah 80

Vermont 44

Virginia 52

Washington 84

Washington, DC 52

West Virginia 60

Wisconsin 66

Wyoming 80

The regional summaries in the pages thatfollow are presented in a geographical order.This alphabetical index will assist you infinding the summary for a particular region.

Species appearing in boldface in the region-al summaries are deemed “unusual” by theregional editors, unless otherwise noted.